Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Prediction of breast cancer sensitivity to neoadjuvant chemotherapy based on status of DNA damage repair proteins

Prediction of breast cancer sensitivity to neoadjuvant chemotherapy based on status of DNA damage... Introduction: Various agents used in breast cancer chemotherapy provoke DNA double-strand breaks (DSBs). DSB repair competence determines the sensitivity of cells to these agents whereby aberrations in the repair machinery leads to apoptosis. Proteins required for this pathway can be detected as nuclear foci at sites of DNA damage when the pathway is intact. Here we investigate whether focus formation of repair proteins can predict chemosensitivity of breast cancer. Methods: Core needle biopsy specimens were obtained from sixty cases of primary breast cancer before and 18-24 hours after the first cycle of neoadjuvant epirubicin plus cyclophosphamide (EC) treatment. Nuclear focus formation of DNA damage repair proteins was immunohistochemically analyzed and compared with tumor response to chemotherapy. Results: EC treatment induced nuclear foci of gH2AX, conjugated ubiquitin, and Rad51 in a substantial amount of cases. In contrast, BRCA1 foci were observed before treatment in the majority of the cases and only decreased after EC in thirteen cases. The presence of BRCA1-, gH2AX-, or Rad51-foci before treatment or the presence of Rad51-foci after treatment was inversely correlated with tumor response to chemotherapy. DNA damage response (DDR) competence was further evaluated by considering all four repair indicators together. A high DDR score significantly correlated with low tumor response to EC and EC + docetaxel whereas other clinicopathological factors analyzed did not. Conclusions: High performing DDR focus formation resulted in tumor resistance to DNA damage-inducing chemotherapy. Our results suggested an importance of evaluation of DDR competence to predict breast cancer chemosensitivity, and merits further studying into its usefulness in exclusion of non-responder patients. Introduction tailoring treatment based on gene status significantly Recent advances in chemotherapy have significantly optimizes the response rate of hormone therapy and improved the prognosis of breast cancer patients. How- trastuzumab, respectively. Prediction of chemosensitivity ever, prediction of tumor sensitivity to chemotherapy with equivalent accuracy is currently anticipated to has not reached a high level of confidence, whereas further improve breast cancer prognosis. determining sensitivity to hormone therapy or trastuzu- Anthracycline-based regimens, such as epirubicin plus mab is relatively more established. Estrogen receptor cyclophosphamide (EC), and taxanes represent the (ER), progesterone receptor (PR) and human epidermal major chemotherapeutic agents used in the breast can- growth factor receptor (HER)2/ErbB2 are practical cer field [1,2]. Of these, anthracycline-based chemother- benchmarks to exclude non-responding patients, and apy induces DNA double-strand breaks (DSBs) [3,4], the most cytotoxic DNA lesion, that leads cells into apoptosis especially when relevant repair pathways * Correspondence: to@marianna-u.ac.jp 1 (represented by homologous recombination (HR) repair) Division of Breast and Endocrine Surgery, Department of Surgery, St. Marianna University School of Medicine, Kawasaki, 216-8511 Japan © 2010 Asakawa et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 2 of 11 http://breast-cancer-research.com/content/12/2/R17 are perturbed [5]. It is important to note that DNA response to DNA damage-inducing chemotherapy and damage repair competence varies among individual whether it correlates with tumor fates after chemo- breast tumors and closely correlates with chemosensitiv- therapy, we analyzed foci in core needle biopsy speci- ity. For example, secondary mutations of BRCA1 or mens from breast cancer before and after neoadjuvant BRCA2 (essential factors in the HR pathway) caused by EC treatment. chemotherapy using cisplatin or poly(ADP-ribose) poly- merase inhibitor in BRCA1/2-mutated cancers restore Materials and methods the wild-type reading frame and, therefore, the tumor Patients and tumors acquires resistance to these drugs [6-8]. These facts Sixty patients with primary breast cancer (2 cm or lar- indicate that chemosensitivity of BRCA-associated can- ger) who consecutively underwent neoadjuvant che- cers could be strongly affected by DNA damage repair motherapy with EC followed by treatment with capability. Based on this evidence it has been suggested docetaxel (DOC) at the Division of Breast and Endo- that HR competence could be a potential biomarker for crine Surgery, St. Marianna University School of Medi- chemosensitivity [9]. Rad51, a protein that plays a direct cine, Japan, were enrolled in the present study from role in HR, especially reflects the HR competence of August 2005 to July 2007. Tumor specimens were cells. Therefore, knowing its status is likely to be valu- obtained by core needle biopsy prior to starting therapy able when assessing HR competence in tumor cells in and 18 to 24 hours after the first cycle of EC treatment. order to instruct therapeutic decisions [9]. Informed consent for the additional core needle biopsy The HR pathway for DSB repair is executed by and experimental use of tumor samples was obtained sequential recruitment of repair proteins to chromatin for all patients in accordance with an approved Institu- around DNA lesions. Accumulation of the proteins is tional Review Board application (registration number regulated by complex mechanisms that utilize phosphor- 946). ylation and ubiquitination modifications mediated by The chemotherapy regimen consisted of four 21-day 2 2 kinases including ataxia telangiectasia mutated (ATM), cycles of EC (E: 80 mg/m on day1,C:600 mg/m on and at least four ubiquitin E3 ligases, RNF8, RNF168, day 1) followed by four 21-day cycles of DOC (75 mg/ 2 2 Rad18, and BRCA1 [10-17]. The Mre11-Rad50-Nbs1 m on day 1). 75 mg/m DOC was administrated four complex first recognizes DSBs and recruits ATM. ATM times as total (only on day 1). There was no increase or then phosphorylates the histone variant H2AX (gH2AX) decrease of the dose. Tumor size was evaluated by [18,19] that triggers accumulation of the downstream E3 three-dimensional images obtained by helical computed ligases RNF8 [11-13,20] and RNF168 [14,15]. Lysine 63- tomography CT scan with a teleradiologic image work- linked polyubiquitin chains built at the sites of DNA station (ZIOSTATION®, Ziosoft Inc., Tokyo, Japan) at damage by these E3 ligases next recruits the BRCA1- baseline, 14 to 21 days after the last cycle of EC, and 21 Abraxas-RAP80 complex through the RAP80 compo- days after the last cycle of DOC treatment. The effect of nent, a protein that contains ubiquitin interacting motif chemotherapy on the tumor was assessed as the three- domains [21-23]. BRCA1 is then essential in order to dimensional volume reduction rate or tumor response rate. recruit repair effector proteins, including Rad51, that The tumor response was evaluated either by Response Eva- perform HR through sister chromatid exchange [24,25]. luation Criteria in Solid Tumors (RECIST) [26] or by the Depletionofany oneofthese proteins resultsinHR three-dimensional volume evaluation defined as: complete deficiency accompanied by loss of Rad51 focus forma- response (CR; disappearance of the disease), partial tion, causing cells to become hypersensitive to DSB- response (PR; reduction of tumor volume of ≥65%), stable inducing agents. disease (SD; volume reduction <65% or enlargement In this study we attempt to clarify the value of HR ≤73%), or progressive disease (PD; volume enlargement competence for the prediction of breast cancer chemo- ≥73%). These are equivalent to CR (disappearance), PR sensitivity. One contention is that nuclear focus forma- (reduction of ≥30%), SD (reduction <30% or enlargement tion of repair proteins in baseline breast cancer tissues ≤20%), or PD (enlargement ≥20%) in unidimensional is a response to spontaneous DNA damage during cell RECIST criteria, respectively (reviewed in [27]). We also proliferationand,inturn, mayrepresent amarkerof analyzed responses with a 50% border between PR and SD HR competence of cells to exogenous DNA damage. (instead of 65%) to evaluate more resistant cases. Therefore, it may predict tumor response to DNA damage-inducing chemotherapy such as with EC. Also, Immunohistochemical analysis the focus formation after chemotherapy could provide Immunohistochemical analysis was performed by us with additional information regarding the DNA the DAKO EnVision system (DAKO, Copenhagen, damage-response capacity. To verify in vivo whether Denmark) with modifications. Formalin-fixed, paraffin- focus formation of repair proteins actually occurs in embedded specimens were cut and heated in a water Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 3 of 11 http://breast-cancer-research.com/content/12/2/R17 bath (95°C, 40 minutes) in Target Retrieval Solution Statistical analysis (pH 9.0, Dako, Carpinteria, CA, USA) for detection of The variables measured in the study were first investi- BRCA-1 or in 10 mM sodium citrate buffer (pH 6.0) gated for association by the chi-squared contingency for gH2AX and Rad51. No pre-treatment was neces- table analysis. For rank correlation, Spearman’s method sary to detect conjugated ubiquitin. After quenching of was performed to determine the correlation between the foci score of two repair proteins and to determine the endogenous peroxidase, the sections were incubated correlation between tumor response rate and focus for- overnight at 4°C with primary antibody at the appro- mation of each repair protein or DDR score. For para- priate dilution [Additional file 1], washed with PBS, metric analyses of tumor volume reduction, Student’s and incubated with horseradish peroxidase-labeled polymer conjugated secondary antibody (EnVision+ unpaired t-test and the Tukey-Kramer method were System, Dako, Carpinteria, CA, USA) for 30 minutes at performed for two-factor comparisons and multiple room temperature. Color development was achieved by comparisons, respectively. For evaluation of significance 3, 3’-diaminobenzidine tetrahydrochloride. Effective- of DDR score and other clinicopathological factors in ness and specificity of each antibody for the detection correlation with mean tumor volume reduction or of DNA damage-induced nuclear foci were verified tumor response rate, variant analysis (univariate) or with cultured cells treated with ionizing radiation (IR) logistic regression analyses (univariate and multivaliate), or epirubicin. The immunofluorescent study has been respectively, were performed. All analyses were carried previously described [28,29]. The nuclear foci were out using Statview 5 statistical software (SAS Institute further analyzed with the protocol used in the tissue Inc, Cary, NC, USA). Statistical significance was stain. The intrinsic subtype[30] was approximated by declared for P values less than 0.05. receptor status determined by standard immunohisto- chemical and fluorescence in situ hybridization (FISH) Results analyses: luminal A: ER+ and/or PR+ and HER2-; Clinical and pathologic features luminal B: ER+ and/or PR+ and HER2+; HER2: ER- Sixty patients with primary breast cancer were included and PR- and HER2+; triple negative: ER- and PR- and in the present series. All tumors were diagnosed as inva- HER2-. Tumors that were immunochistochemically sive ductal carcinoma. Patient clinical characteristics are scored as 3+, or 2+ with FISH-positive, were regarded given in Table 1. All triple-negative tumors were posi- tive for CK5/6 (therefore described as basal-like in as positive for HER2 status. Cytokeratin (CK) 5/6 Table 1) whereas three cases of Luminal A, one case of expression was also examined to evaluate the basal-like Luminal B and three cases of HER2 type were positive character. for CK5/6. Three patients have one first-degree relative Immunohistochemical scoring with a history of breast cancer and two patients have Taking into consideration that all immunohistochem- one second-degree relative with a history of breast or ical markers used in the study localize to sites of DNA ovarian cancer. All patients completed an EC plus DOC damage in the normal HR pathway, we only counted cells displaying nuclear focus formation and disre- garded cytoplasmic or diffuse nuclear staining. We Table 1 Patient characteristics scored the nuclear foci staining as follows: 0 = no posi- Characteristic Number of Characteristic Number of tive cells, 1 = less than 10% positive cells, 2 = 10% or factor patients factor patients greater, but less than 80% positive cells, 3 = 80% or Age at treatment Cancer stage greater positive cells. Two observers (HA and HK) start were blinded to the clinical information to avoid Median 50 I 0 observer subjectivity when evaluating the immunohis- Range 34-68 II 53 tochemical staining. To correlate staining with tumor Lymph node metastasis III 5 response, we divided the cases into negative and posi- Negative 36 IV 2 tive samples to simplify the statistical analyses. The Positive 24 Intrinsic subtype* positive cases are a total of the categories with a foci Tumor stage Luminal A 37 score of 1, 2 and 3. To assess the capacity of the DNA T1 0 Luminal B 6 damage response (DDR) using a more comprehensive T2 54 HER2 11 approach, we configured the DDR score by counting T3 6 Basal-like 6 the total number of positive factors present in baseline T4 0 Total 60 foci of BRCA1, gH2AX and Rad51, and EC-induced * intrinsic subtypes were approximated by immunohistochemical receptor status. HER, human epidermal growth factor receptor. foci of Rad51, per case. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 4 of 11 http://breast-cancer-research.com/content/12/2/R17 regimen. Rad51 and gH2AX stains were not performed BRCA1 after EC (EC-induced foci score) significantly on tumor specimens before EC in two patients because correlated with that of Rad51 (P = 0.0017; Table 2), of insufficient tumor sample after reserving stocks for likely reflecting the requirement of BRCA1 for Rad51 clinical use. Tumor size evaluation by CT after treat- recruitment at the site of DNA damage. However, no ment with EC plus DOC was not performed for one other correlations between repair proteins were patient because of the patient’s condition. All but one observed, and no clear pattern combinations of repair patient received breast surgery after EC and DOC. proteins emerged. Nuclear foci staining of DNA damage repair proteins Association of focus formation of each repair protein with To assess the competence of the DSB repair pathway, tumor response to chemotherapy we immunohistochemically analyzed gH2AX, conjugated To elucidate the possible association between DDR com- ubiquitin, BRCA1, and Rad51 in nuclear foci based on petence and tumor response to chemotherapy, we corre- the idea that these candidates may represent a typical lated the presence of individual repair proteins in foci course of the DSB repair cascade [31]. Of these, gH2AX with tumor volume before and after chemotherapy. is the most upstream element, sequentially followed in Tumor volume was measured prior to chemotherapy to the cascade by conjugated ubiquitin, BRCA1, and establish the baseline volume. The mean volume reduc- Rad51. Rad51 is the most downstream of these four pro- tion of tumors after EC and after EC and DOC was 59.7 teins and is directly involved in HR. However, it should ± 25.8% and 76.0 ± 20.7% of baseline tumor volume, be mentioned that DNA repair failure due to genes at respectively. We analyzed the presence of repair proteins the same level of or downstream of RAD51, such as in foci before (baseline foci) and after EC treatment (EC- RAD51AP1 [32] or translesion DNA polymerases induced foci), sorted them into positive and negative foci [33,34], is an unlikely cause of loss of foci formation of groups for each individual repair protein, and then corre- these proteins. In addition to untreated, baseline breast lated each group with tumor volume [Additional file 2]. cancer tissues, we analyzed the tissues 18 to 24 hours There was a significant difference in tumor volume after after the first cycle of EC treatment to obtain further EC between BRCA1-positive and BRCA1-negative base- information for the assessment of DNA repair capacity. line foci groups (82.1 ± 17.8% vs 55.7 ± 25.1%, P = Theantibodiesusedinthisstudy arecommonlyused 0.0039) [Additional file 2a]. We then performed the same and well characterized in general. In addition we tested analysis after EC and DOC treatment. In addition to background staining and confirmed the specific detec- BRCA1 (93.7 ± 6.6% vs 72.8 ± 20.7%, P = 0.0044), signifi- tion of nuclear foci at DSBs caused by IR or epirubucin cant differences in tumor volume were observed between treatment (Figure 1). positive and negative gH2AX (78.4 ± 17.4% vs 65.6 ± The immunohistochemical analyzes revealed that in all 26.8%, P = 0.0429) and Rad51 baseline foci groups (78.1 but two cases, the foci score of at least one of the repair ± 18.9% vs 63.6 ± 24.4%, P = 0.0351) [Additional file 2b]. proteins was altered in response to EC treatment. We next tested the correlation between scored foci Representative data for immunohistochemical findings groups and the tumor response rate. The tumor of the nuclear focus formation of the repair proteins response rate was evaluated with RECIST or three- before and after the first cycle of EC are shown in dimensional volume reduction using either 65% or 50% Figure 2 with panels summarizing the foci scores of the of the PR/SD border (as described in the Materials and cases. Prior to EC treatment, samples were stained to Methods). Tumor responses to EC and EC plus DOC determine baseline staining of foci. The foci were posi- according to focus formation status are shown in Addi- tive for gH2AX (20 of58cases), BRCA1(51 of 60 tional file 3. Contingency table analyses demonstrate sig- cases), or Rad51 (11 of 58 cases) whereas no cases nificant differences in the EC tumor response rate exhibited foci staining for conjugated ubiquitin (0 of 60 between BRCA1-positive and BRCA1-negative baseline cases). In response to EC treatment, the number of foci foci groups and between Rad51-positive and Rad51- staining positive for gH2AX (44 of 58 cases), conjugated negative EC-induced foci groups for all three criteria of ubiquitin (26 of 60 cases), and Rad51 (31 of 58 cases) the response rate (ZIO 65%, ZIO 50%, RECIST). There increased, whereas foci staining for BRCA1 either continued to be a significant difference in tumor increased (9 of 60 cases), remained unchanged (38 of 60 response rate after EC and DOC treatment between cases) or decreased (13 of 60 cases). The reason why Rad51-positive and Rad51-negative EC-induced foci BRCA1 foci staining decreased after treatment in some groups for all three criteria. In addition, when evaluated cases is not clear at present but it could be implicated with a three-dimensional volume reduction using 50% of in the presence of BRCA1 foci in normal S-phase that the PR/SD border, significant differences in the tumor colocalizes with proliferating cell nuclear antigen response rate to EC and DOC were observed between (PCNA) at DNA replication fork [35]. The foci score of Rad51-positive and Rad51-negative baseline foci groups. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 5 of 11 http://breast-cancer-research.com/content/12/2/R17 Figure 1 Immunohistochemistry controls and antibody specificity. (a) Immunohistochemical staining with control IgG for tumors after the first cycle of epirubicin plus cyclophosphamide (EC) treatment (upper panel). Lower panel shows Rad51 staining for morphologically diagnosed non-cancerous breast tissues (left upper part) and tumor (right lower part) after the first cycle of EC treatment. Although non-cancerous breast cells also expressed nuclear foci formation the number and intensity was significantly lower than that in tumor cells. (b and c) DNA damage- induced nuclear foci formations detected by antibodies used in the study. HeLa cells were either untreated (-), treated with 5 Gy ionizing radiation (IR) or 0.2 μg/ml epirubicin (Epi), incubated for three hours and fixed. Cells were then subjected either to immunofluorescent analyses with the indicated primary antibodies and (b) FITC- (green) or Rhodamine- (red) conjugated secondary antibodies, or (c) to the same protocol as that used in the tissue stain. For immunofluorescent analyses the nucleus was counterstained with DAPI. Ub, conjugated ubiquitin. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 6 of 11 http://breast-cancer-research.com/content/12/2/R17 Figure 2 Nuclear focus formation in response to chemotherapy. Tumor specimens were obtained by core needle biopsy before and 18 to 24 hours after the first cycle of epirubicin plus cyclophosphamide (EC) treatment. Immunohistochemical findings from representative cases for gH2AX, conjugated ubiquitin (Ub), BRCA1, and Rad51 are shown. Graphs at right demonstrate changes in nuclear foci score after EC treatment in all cases analyzed. The red, blue, and black lines indicate cases with increased, decreased, and unchanged scores, respectively. The thickness of the lines proportionally reflects the number of cases. The thinnest line (gH2AX score 3 to 3) and the thickest line (Ub, score 0 to 0) represent 1 and 34 cases, respectively. n, number of cases analyzed. To specify the correlation of these focus formation Table 2 Correlation between EC-induced foci of Rad51 and BRCA1 groups with tumor response rates, we further analyzed thedatawithSpearman ’s rank correlation method. BRCA1 When evaluated with three-dimensional volume reduc- 0 1 2 3 Total tion using 50% of the PR/SD border, Spearman’sanaly- Rad51 sis showed that the presence of BRCA1-positive baseline 0 7 10 2 0 19 foci associated with poor EC tumor response (P = 1 3 24 9 2 38 0.0067) [Additional file 3]. Spearman’sanalysisalso 20 0 2 0 2 demonstrated that the presence of Rad51-positive base- Total 10 34 13 2 59 line foci (P = 0.0078) or EC-induced foci (P = 0.0042) P = 0.0017 [Additional file 3] associated with poor EC and DOC P value is from the Spearman’s rank correlation test. EC, epirubicin plus tumor response. cyclophosphamide. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 7 of 11 http://breast-cancer-research.com/content/12/2/R17 Association of DDR score with tumor reduction by chemotherapy The analysis correlating focus formation of BRCA1, gH2AX, and Rad51 prior to treatment and of Rad51 foci after EC treatment with the mean tumor volume reduc- tion or tumor response rate [Additional files 2 and 3] uncovers a significant inverse correlation with tumor response for each of the four conditions. These data sup- port the supposition that higher DDR competency pro- duces tumors resistant to chemotherapy. To correlate overall DDR competency with tumor reduction, we devised a simple measurement to assess DDR competency. Each patient case was analyzed for the presence of all four of the above listed conditions and was assigned a DDR score of 0 to 4 based on the number of conditions present. This DDR score was then correlated with mean tumor volume reductions. Number of cases in each DDR score is shown in Table 3. As shown in Figure 3, both the mean tumor volume reductions after EC (28.4 ± 28.1%) and afterECand DOC (49.9 ±22.0%) for DDR score4(all four conditions present) were the lowest among all the scores. There were significant differences between score 4 and either score 0 or 2 for the mean tumor volume reduc- tions after EC (Figure 3a) and between score 4 and either score 0, 1 or 2 for the mean tumor volume reductions after EC and DOC (Figure 3b), as judged by the Tukey- Kramer multiple comparisons study setting P <0.05asa significance threshold. In addition, Spearman’sanalysis showed that a high DDR score was associated with poor tumor response rate after EC and DOC (P = 0.0031) when evaluated with three-dimensional volume reduction using 50% of the PR/SD border (Table 3). A high DDR score also tended to be associated with poor tumor response rate after EC (P = 0.0639, Table 3). Table 3 Correlation between DDR score and tumor response rate (ZIO 50%) Figure 3 Mean tumor volume reductions after (a) EC or (b) EC DDR score and DOC according to DDR score. Error bars represent standard 0 1 2 3 4 Total deviation. Significance was analyzed by Tukey-Kramer test setting P < 0.05 as significance threshold. DDR, DNA damage response; DOC, After EC docetaxel; EC, epirubicin plus cyclophosphamide. CR 1 0 1 0 0 2 PR 1 12 14 9 2 38 SD 0 3 8 3 4 18 The correlation between DDR score and the tumor Total 2 15 23 12 6 58 response prompted us to examine whether it has a sig- P = 0.0639 nificant impact among other clinicopathological factors After EC + DOC including age, cancer stage, tumor size, nodal metastatic CR 1 1 3 1 0 6 status, and subtypes. None of these factors correlated PR 1 13 19 9 2 44 with DDR score (data not shown). The variant analysis SD 0 0 1 2 4 7 for mean tumor volume reduction after EC revealed Total 2 14 23 12 6 57 that only DDR score (P = 0.0069), but no other factors P = 0.0031 correlated with the mean tumor volume reduction P values are from the Spearman’s rank correlation test. CR, complete response; (Table 4). The variant analysis for mean tumor volume DDR, DNA damage response; EC, epirubicin plus cyclophosphamide; PR, partial response; SD, stable disease. reduction after EC and DOC also demonstrated DDR Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 8 of 11 http://breast-cancer-research.com/content/12/2/R17 score (P = 0.0035) as the most significant correlation patients and we also considered the patient’sconveni- factor, followed by nodal status (P = 0.0201) and tumor ence. However, the ideal timing remains to be deter- size (P = 0.0538, Table 4). In addition, univariate logistic mined if biopsy after chemotherapy is required. regression analysis showed that a high (3 and 4) DDR The second reason for the diversity of the DDR score was most significantly associated with poor tumor response could be attributed to the diversity of aberra- response after EC and DOC (P = 0.0095) when evalu- tions of the genes responsible for DSB repair in each ated with volume reductionusing 50%ofthe PR/SD breast cancer. Theoretically, defects in the recruitment border, followed by tumor size (P = 0.0260), cancer of upstream repair proteins could result in loss of down- stage (P = 0.0465), and subtype (P = 0.0659, Table 5). stream proteins at sites of DNA damage, and this has We then examined multivariate analysis with tumor been shown to be the case in many molecular biological size, nodal status, subtype, and DDR score, the factors studies using cultured cells [10-15,21-23]. Furthermore, that showed probable association with tumor response it was also shown that Rad51 nuclear expression is rate in the univariate analysis. Cancer stage was omitted absent in tumors associated with BRCA2 mutation [41]. because it was correlated with tumor size. Importantly Thepositivecorrelation foundbetween EC-induced the result indicated that only the DDR score was signifi- BRCA1 and Rad51 foci in this study (Table 2) may also cantly associated with tumor response rate (P =0.0402) support this interpretation. In contrast, it was reported independent of other factors analyzed (Table 6). that overexpression of Rad51 restored Rad51 focus for- mation and rescued the sensitivity of BRCA1-deficient Discussion cells to x-rays and cisplatin [42]. Importantly, up-regula- In the present study using human tumor specimens we tion of Rad51 was a common feature of BRCA1-defi- show for the first time that DNA repair competence may cient breast tumors [42]. These data suggest that the predict breast cancer sensitivity to DNA damage-inducing mechanism of DSB repair response in vivo is not simple chemotherapy. We selected gH2AX, conjugated ubiquitin, and that assessment of DSB repair aberrations in each BRCA1, and Rad51, proteins in the DSB repair cascade, to patient case is, therefore, unreasonable at present. assess DNA repair competence because accumulated evi- In an attempt to address this problem in our current dence demonstrates that inactivation of genes in the DSB study, we assessed the comprehensive capacity of DSB repair pathway results in cellular sensitivity to DNA repair by incorporating multiple candidate factors into damage-inducing chemotherapy [16,29,31,36-38]. In our one DDR score. We found that foci of BRCA1, gH2AX, study, these repair proteins dramatically responded to EC and Rad51 prior to treatment and EC-induced foci of treatment. The conjugated ubiquitin response was espe- Rad51 correlated with tumor response when compared cially dramatic as approximately half of the cases analyzed either with the mean tumor volume reduction or the formed conjugated ubiquitin foci, compared with unde- tumor response rate. When incorporating these four fac- tectable foci formation prior to treatment. This suggests tors into one DDR score a significant correlation was that ubiquitination occurs in vivo during the DNA damage observed with mean tumor volume reduction after EC, response in an early stage after chemotherapy. However, whereas no other factors correlated with the mean in spite of the dramatic response, we did not find any sig- tumor volume reduction (Table 4 and Figure 3a). nificant correlation between conjugated ubiquitin foci for- Although it was not statistically significant, the similar mation and tumor response. The reason is currently correlation was also observed between DDR score and unknown. One possibility is that this could be attributed tumor response rate (Table 3). These correlations to the fact that ubiquitination is also involved in DNA became more significant after EC and DOC treatment damage response pathways other than for DSBs. (Tables 3 to 5 and Figure 3b) and the DDR score was We did not find certain trends of the combinations of an independent predictive factor of other factors includ- responding repair proteins. Several reasons could ing tumor subtype when evaluated with volume reduc- account for this observation. First, the metabolism and tion using 50% of the PR/SD border (Table 6). Recent pharmacokinetics of the agents could vary per patient. studies suggested that luminal tumors have low response The ideal time to obtain the in vivo sample was, there- rate to neoadjuvant chemotherapy, whereas basal-like fore, difficult to determine. The experimental design and HER2+ tumors have higher response rates. For employed in this study was not very robust in this way. example, it has been reported that clinical response rate In cultured cells, gH2AX accumulates at sites of DNA (CR and PR) to anthracyclin-based chemotherapy of damage just minutes after the damage occurs, whereas luminal A was 39%, whereas that of basal-like, which BRCA1 and Rad51 foci appear 30 minutes to several has been implicated with BRCA1 dysfunction [43,44], hours afterwards [11,35,39,40]. In this study we har- was 85% [45]. The response rates to EC treatment of vested samples 18 to 24 hours after EC treatment luminal A (15 of 37 cases, 40.5%) and basal-like (4 of 6 because the agents were still expected to be present in cases, 66.7%) subtypes in the current study were not Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 9 of 11 http://breast-cancer-research.com/content/12/2/R17 Table 4 Univariate analysis of variance for mean tumor Table 4: Univariate analysis of variance for mean tumor volume reduction volume reduction (Continued) n Mean tumor volume P 3 12 70.7 ± 27.5 reduction (%) ± SD 4 6 49.9 ± 22.0 After EC DDR, DNA Damage Response; DOC, docetaxel; EC, epirubicin plus Age (years) cyclophosphamide; HER, human epidermal growth factor receptor; SD, standard deviation. -50 30 64.1 ± 25.9 0.1710 51- 29 54.9 ± 25.4 Cancer stage Table 5 Univariate logistic regression analysis of factors II 53 61.5 ± 25.1 0.0962 affecting tumor response rate (ZIO 50%) III 5 55.7 ± 29.1 Odds ratio (95% CI) P IV 2 21.9 ± 6.4 After EC + DOC Tumor stage Age (years) T2 54 60.8 ± 25.3 0.3059 <51 1.000 T3 6 21.9 ± 6.4 51≤ 0.800 (0.192-3.333) 0.7592 Nodal status Cancer stage N- 24 62.7 ± 24.9 0.4557 II 1.000 N+ 36 57.6 ± 26.6 III, IV 5.750 (1.028-32.174) 0.0465 Subtype Tumor stage Luminal A 37 58.8 ± 21.3 0.2923 T2 1.000 Luminal B 6 72.9 ± 9.4 T3 7.833 (1.279-47.964) 0.0260 HER2 11 50.4 ± 36.4 Nodal status Basal-like 6 69.4 ± 37.0 N- 1.000 DDR score N+ 6.286 (0.730-54.110) 0.0942 0 2 94.6 ± 7.6 0.0069 Subtype 1 15 57.3 ± 19.4 Luminal A, B 1.000 2 23 62.8 ± 22.9 HER2, Basal-like 3.958 (0.913-17.154) 0.0659 3 12 61.1 ± 26.6 DDR score 4 6 28.4 ± 28.1 0, 1, 2 1.000 After EC-DOC 3, 4 9.423 (1.729-51.359) 0.0095 Age CI, confidence interval; DDR, DNA Damage Response; DOC, docetaxel; EC, -50 30 80.6 ± 23.9 0.0804 epirubicin plus cyclophosphamide; HER, human epidermal growth factor receptor. 51- 29 71.2 ± 15.7 Cancer stage very different from the previous report. However, we II 52 77.9 ± 19.8 0.1230 could not find any correlation between subtype and III 5 64.3 ± 25.7 DDR score while DDR score independently predicted IV 2 54.7 ± 17.6 the chemosensitivity. The result may reflect the fact that Tumor stage luminal A tumors also include DNA damage-sensitive T2 53 77.7 ± 19.7 0.0538 tumors with defective HR pathways that can be counted T3 6 60.6 ± 24.7 by the DDR score. Supporting this it has been shown Nodal status that tumors caused by BRCA2 deficiency mainly become N- 23 83.7 ± 14.3 0.0201 luminal A tumors [44,46,47]. N+ 36 71.0 ± 22.7 The reason why the correlation between the DDR score Subtype and tumor response after EC and DOC treatment became Luminal A 36 77.9 ± 16.1 0.0789 more significant than that after EC is not clear at present. Luminal B 6 86.0 ± 9.5 As DOC does not induce DNA DSBs, the observed effect HER2 11 62.4 ± 30.5 is not likely to be due to the sensitivity to DNA damage in Basal-like 6 79.7 ± 24.8 those tumors. DOC might be more toxic for the cells with DDR score gross genomic aberration caused by the pretreatment with 0 2 99.6 ± 0.6 0.0035 EC under the condition of being less HR competent. 1 14 77.2 ± 13.2 Alternatively it is possible that time length after EC treat- 2 23 81.0 ± 14.7 ment enhanced the difference of the outcome. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 10 of 11 http://breast-cancer-research.com/content/12/2/R17 epirubicin plus cyclophosphamide; ER: estrogen receptor; HER: human Table 6 Multivariate logistic regression analysis of factors epidermal growth factor receptor; HR: homologous recombination repair; IR: affecting tumor response rate (ZIO 50%) ionizing radiation; PBS: phosphate-buffered saline; PD: progressive disease; Odds ratio (95% CI) P PR: partial response; PR (+ or -): progesterone receptor (+ or -); RECIST: Response Evaluation Criteria in Solid Tumors; SD: stable disease. After EC + DOC Tumor stage Acknowledgements T2 1.000 This study was supported by grants from the Ministry of Education, Science, Sports, Culture and Technology of Japan grants-in-aid. We thank Mrs. T3 2.246 (0.290-17.420) 0.4388 Shigeko Ohnuma for technical support for immunohistochemical Nodal status experiments; Dr. Yasuo Miyoshi for technical advice; Dr. Ettore Appella for N- 1.000 critical reading of the manuscript; Members in the Division of Breast and Endocrine Surgery for support in collection of biopsy specimens; and Mrs. N+ 3.651 (0.346-38.506) 0.2813 Rie Ogawa and Miss Umi Noguchi for secretarial assistance. Subtype Luminal A, B 1.000 Author details Division of Breast and Endocrine Surgery, Department of Surgery, St. HER2, Basal-like 2.484 (0.464-13.287) 0.2874 Marianna University School of Medicine, Kawasaki, 216-8511 Japan. DDR score Department of Translational Oncology, St. Marianna University Graduate 0, 1, 2 1.000 School of Medicine, Kawasaki, 216-8511 Japan. Department of Diagnostic Pathology, St. Marianna University School of Medicine, Kawasaki, 216-8511 3, 4 6.694 (1.088-41.182) 0.0402 Japan. Department of Breast and Endocrine Surgery, Kumamoto University, CI, confidence interval; DDR, DNA Damage Response; DOC, docetaxel; EC, Honjo 1-1-1, Kumamoto 860-8556, Japan. epirubicin plus cyclophosphamide; HER, human epidermal growth factor receptor. Authors’ contributions HA analyzed the majority of the data. HK conducted immunohistochemical analyses. AK obtained the data for tumor response. MT supported Interestingly, DDR score group 4 consisted of cases immunohistochemical analyses. WW characterized antibody specificities. HI with poor tumor responses to chemotherapy when eval- and MF made substantial contributions to analysis and interpretation of the data. TO designed and conducted the studies, and wrote the manuscript. All uated for both mean tumor volume reduction (Figure 3) authors read and approved the final manuscript. and tumor response rate (Table 3). This result may lead to the possibility of using DDR status in the clinic to Competing interests The authors declare that they have no competing interests. predict and exclude non-responders to EC treatment. It is noteworthy to point out that the HR repair cascade Received: 18 September 2009 Revised: 7 January 2010 for DSB contains many essential proteins other than Accepted: 5 March 2010 Published: 5 March 2010 those tested in this study. By including select subsets of References proteins for analysis, it may be possible to identify non- 1. Trudeau M, Charbonneau F, Gelmon K, Laing K, Latreille J, Mackey J, responders in order to avoid unnecessary chemotherapy. McLeod D, Pritchard K, Provencher L, Verma S: Selection of adjuvant Ideally in such cases, the levels of baseline foci present chemotherapy for treatment of node-positive breast cancer. Lancet Oncol 2005, 6:886-898. prior to treatment would provide enough information to 2. Effects of chemotherapy and hormonal therapy for early breast cancer determine appropriate treatment, preventing the need on recurrence and 15-year survival: an overview of the randomised for additional core needle biopsy after chemotherapy. trials. Lancet 2005, 365:1687-1717. 3. Capranico G, Zunino F, Kohn KW, Pommier Y: Sequence-selective topoisomerase II inhibition by anthracycline derivatives in SV40 DNA: Conclusions relationship with DNA binding affinity and cytotoxicity. Biochemistry 1990, In conclusion, our results suggest the importance of 29:562-569. 4. Capranico G, De Isabella P, Penco S, Tinelli S, Zunino F: Role of DNA evaluating DDR competence to predict breast cancer breakage in cytotoxicity of doxorubicin, 9-deoxydoxorubicin, and chemosensitivity and warrant further investigation into 4-demethyl-6-deoxydoxorubicin in murine leukemia P388 cells. Cancer itseffectivenessasawaytoexclude non-responding Res 1989, 49:2022-2027. 5. Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K, Linn S: Molecular mechanisms patients. of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004, 73:39-85. Additional file 1: Table S1. Antibodies used in the present 6. Sakai W, Swisher EM, Karlan BY, Agarwal MK, Higgins J, Friedman C, immunohistochemical study. Villegas E, Jacquemont C, Farrugia DJ, Couch FJ, Urban N, Taniguchi T: Secondary mutations as a mechanism of cisplatin resistance in BRCA2- Additional file 2: Figure S1. Mean tumor volume reduction after EC (a) mutated cancers. Nature 2008, 451:1116-1120. or EC+DOC (b) according to the nuclear foci status for DNA repair 7. Edwards SL, Brough R, Lord CJ, Natrajan R, Vatcheva R, Levine DA, Boyd J, proteins. Reis-Filho JS, Ashworth A: Resistance to therapy caused by intragenic Additional file 3: Table S2. Tumor response rate according to the deletion in BRCA2. Nature 2008, 451:1111-1115. nuclear foci status for DNA repair proteins. 8. Swisher EM, Sakai W, Karlan BY, Wurz K, Urban N, Taniguchi T: Secondary BRCA1 mutations in BRCA1-mutated ovarian carcinomas with platinum resistance. Cancer Res 2008, 68:2581-2586. 9. Livingston DM, Silver DP: Cancer: crossing over to drug resistance. Nature Abbreviations 2008, 451:1066-1067. CK: cytokeratin; CR: complete response; CT: computed tomography; DDR: DNA damage response; DOC: docetaxel; DSB: DNA double-strand break; EC: Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 11 of 11 http://breast-cancer-research.com/content/12/2/R17 10. Wang B, Elledge SJ: Ubc13/Rnf8 ubiquitin ligases control foci formation expression patterns of breast carcinomas distinguish tumor subclasses of the Rap80/Abraxas/Brca1/Brcc36 complex in response to DNA with clinical implications. Proc Natl Acad Sci USA 2001, 98:10869-10874. damage. Proc Natl Acad Sci USA 2007, 104:20759-20763. 31. Harper JW, Elledge SJ: The DNA damage response: ten years after. Mol 11. Mailand N, Bekker-Jensen S, Faustrup H, Melander F, Bartek J, Lukas C, Cell 2007, 28:739-745. Lukas J: RNF8 Ubiquitylates Histones at DNA Double-Strand Breaks and 32. Wiese C, Dray E, Groesser T, San Filippo J, Shi I, Collins DW, Tsai MS, Promotes Assembly of Repair Proteins. Cell 2007, 131:887-900. Williams GJ, Rydberg B, Sung P, Schild D: Promotion of homologous 12. Kolas NK, Chapman JR, Nakada S, Ylanko J, Chahwan R, Sweeney FD, recombination and genomic stability by RAD51AP1 via RAD51 Panier S, Mendez M, Wildenhain J, Thomson TM, Pelletier L, Jackson SP, recombinase enhancement. Mol Cell 2007, 28:482-490. Durocher D: Orchestration of the DNA-damage response by the RNF8 33. Kawamoto T, Araki K, Sonoda E, Yamashita YM, Harada K, Kikuchi K, ubiquitin ligase. Science 2007, 318:1637-1640. Masutani C, Hanaoka F, Nozaki K, Hashimoto N, Takeda S: Dual roles for 13. Huen MS, Grant R, Manke I, Minn K, Yu X, Yaffe MB, Chen J: RNF8 DNA polymerase eta in homologous DNA recombination and translesion Transduces the DNA-Damage Signal via Histone Ubiquitylation and DNA synthesis. Mol Cell 2005, 20:793-799. Checkpoint Protein Assembly. Cell 2007, 131:901-914. 34. McIlwraith MJ, Vaisman A, Liu Y, Fanning E, Woodgate R, West SC: Human 14. Doil C, Mailand N, Bekker-Jensen S, Menard P, Larsen DH, Pepperkok R, DNA polymerase eta promotes DNA synthesis from strand invasion Ellenberg J, Panier S, Durocher D, Bartek J, Lukas J, Lukas C: RNF168 binds intermediates of homologous recombination. Mol Cell 2005, 20:783-792. and amplifies ubiquitin conjugates on damaged chromosomes to allow 35. Scully R, Chen J, Ochs RL, Keegan K, Hoekstra M, Feunteun J, accumulation of repair proteins. Cell 2009, 136:435-446. Livingston DM: Dynamic changes of BRCA1 subnuclear location and 15. Stewart GS, Panier S, Townsend K, Al-Hakim AK, Kolas NK, Miller ES, phosphorylation state are initiated by DNA damage. Cell 1997, Nakada S, Ylanko J, Olivarius S, Mendez M, Oldreive C, Wildenhain J, 90:425-435. Tagliaferro A, Pelletier L, Taubenheim N, Durandy A, Byrd PJ, Stankovic T, 36. Murakawa Y, Sonoda E, Barber LJ, Zeng W, Yokomori K, Kimura H, Niimi A, Taylor AM, Durocher D: The RIDDLE syndrome protein mediates a Lehmann A, Zhao GY, Hochegger H, Boulton SJ, Takeda S: Inhibitors of the ubiquitin-dependent signaling cascade at sites of DNA damage. Cell proteasome suppress homologous DNA recombination in mammalian 2009, 136:420-434. cells. Cancer Res 2007, 67:8536-8543. 16. Wu W, Koike A, Takeshita T, Ohta T: The ubiquitin E3 ligase activity of 37. Collis SJ, Swartz MJ, Nelson WG, DeWeese TL: Enhanced radiation and BRCA1 and its biological functions. Cell Div 2008, 3:1. chemotherapy-mediated cell killing of human cancer cells by small 17. Huang J, Huen MS, Kim H, Leung CC, Glover JN, Yu X, Chen J: RAD18 inhibitory RNA silencing of DNA repair factors. Cancer Res 2003, transmits DNA damage signalling to elicit homologous recombination 63:1550-1554. repair. Nat Cell Biol 2009, 11:592-603. 38. Foulkes WD: BRCA1 and BRCA2: chemosensitivity, treatment outcomes 18. Bassing CH, Suh H, Ferguson DO, Chua KF, Manis J, Eckersdorff M, and prognosis. Fam Cancer 2006, 5:135-142. Gleason M, Bronson R, Lee C, Alt FW: Histone H2AX: a dosage-dependent 39. Ayoub N, Jeyasekharan AD, Bernal JA, Venkitaraman AR: HP1-beta suppressor of oncogenic translocations and tumors. Cell 2003, mobilization promotes chromatin changes that initiate the DNA damage 114:359-370. response. Nature 2008, 453:682-686. 19. Celeste A, Fernandez-Capetillo O, Kruhlak MJ, Pilch DR, Staudt DW, Lee A, 40. Zhong Q, Chen CF, Li S, Chen Y, Wang CC, Xiao J, Chen PL, Sharp ZD, Bonner RF, Bonner WM, Nussenzweig A: Histone H2AX phosphorylation is Lee WH: Association of BRCA1 with the hRad50-hMre11-p95 complex dispensable for the initial recognition of DNA breaks. Nat Cell Biol 2003, and the DNA damage response. Science 1999, 285:747-750. 5:675-679. 41. Honrado E, Osorio A, Palacios J, Milne RL, Sanchez L, Diez O, Cazorla A, 20. Plans V, Scheper J, Soler M, Loukili N, Okano Y, Thomson TM: The RING Syrjakoski K, Huntsman D, Heikkila P, Lerma E, Kallioniemi A, Rivas C, finger protein RNF8 recruits UBC13 for lysine 63-based self Foulkes WD, Nevanlinna H, Benitez J: Immunohistochemical expression of polyubiquitylation. J Cell Biochem 2006, 97:572-582. DNA repair proteins in familial breast cancer differentiate BRCA2- 21. Kim H, Chen J, Yu X: Ubiquitin-binding protein RAP80 mediates BRCA1- associated tumors. J Clin Oncol 2005, 23:7503-7511. dependent DNA damage response. Science 2007, 316:1202-1205. 42. Martin RW, Orelli BJ, Yamazoe M, Minn AJ, Takeda S, Bishop DK: RAD51 up- 22. Sobhian B, Shao G, Lilli DR, Culhane AC, Moreau LA, Xia B, Livingston DM, regulation bypasses BRCA1 function and is a common feature of BRCA1- Greenberg RA: RAP80 targets BRCA1 to specific ubiquitin structures at deficient breast tumors. Cancer Res 2007, 67:9658-9665. DNA damage sites. Science 2007, 316:1198-1202. 43. Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, Deng S, 23. Wang B, Matsuoka S, Ballif BA, Zhang D, Smogorzewska A, Gygi SP, Johnsen H, Pesich R, Geisler S, Demeter J, Perou CM, Lonning PE, Elledge SJ: Abraxas and RAP80 form a BRCA1 protein complex required Brown PO, Borresen-Dale AL, Botstein D: Repeated observation of breast for the DNA damage response. Science 2007, 316:1194-1198. tumor subtypes in independent gene expression data sets. Proc Natl 24. Greenberg RA, Sobhian B, Pathania S, Cantor SB, Nakatani Y, Livingston DM: Acad Sci USA 2003, 100:8418-8423. Multifactorial contributions to an acute DNA damage response by 44. Turner N, Tutt A, Ashworth A: Hallmarks of ‘BRCAness’ in sporadic BRCA1/BARD1-containing complexes. Genes Dev 2006, 20:34-46. cancers. Nat Rev Cancer 2004, 4:814-819. 25. Yu DS, Sonoda E, Takeda S, Huang CL, Pellegrini L, Blundell TL, 45. Carey LA, Dees EC, Sawyer L, Gatti L, Moore DT, Collichio F, Ollila DW, Venkitaraman AR: Dynamic control of Rad51 recombinase by self- Sartor CI, Graham ML, Perou CM: The triple negative paradox: primary association and interaction with BRCA2. Mol Cell 2003, 12:1029-1041. tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res 2007, 26. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, 13:2329-2334. Verweij J, Van Glabbeke M, van Oosterom AT, Christian MC, Gwyther SG: 46. Lakhani SR, Vijver Van De MJ, Jacquemier J, Anderson TJ, Osin PP, New guidelines to evaluate the response to treatment in solid tumors. McGuffog L, Easton DF: The pathology of familial breast cancer: European Organization for Research and Treatment of Cancer, National predictive value of immunohistochemical markers estrogen receptor, Cancer Institute of the United States, National Cancer Institute of progesterone receptor, HER-2, and p53 in patients with mutations in Canada. J Natl Cancer Inst 2000, 92:205-216. BRCA1 and BRCA2. J Clin Oncol 2002, 20:2310-2318. 27. Jaffe CC: Measures of response: RECIST, WHO, and new alternatives. J 47. Foulkes WD, Metcalfe K, Sun P, Hanna WM, Lynch HT, Ghadirian P, Tung N, Clin Oncol 2006, 24:3245-3251. Olopade OI, Weber BL, McLennan J, Olivotto IA, Begin LR, Narod SA: 28. Sato K, Hayami R, Wu W, Nishikawa T, Nishikawa H, Okuda Y, Ogata H, Estrogen receptor status in BRCA1- and BRCA2-related breast cancer: Fukuda M, Ohta T: Nucleophosmin/B23 is a candidate substrate for the the influence of age, grade, and histological type. Clin Cancer Res 2004, BRCA1-BARD1 ubiquitin ligase. J Biol Chem 2004, 279:30919-30922. 10:2029-2034. 29. Takeshita T, Wu W, Koike A, Fukuda M, Ohta T: Perturbation of DNA repair doi:10.1186/bcr2486 pathways by proteasome inhibitors corresponds to enhanced Cite this article as: Asakawa et al.: Prediction of breast cancer sensitivity chemosensitivity of cells to DNA damage-inducing agents. Cancer to neoadjuvant chemotherapy based on status of DNA damage repair Chemother Pharmacol 2009, 64:1039-46. proteins. Breast Cancer Research 2010 12:R17. 30. Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, Hastie T, Eisen MB, Rijn van de M, Jeffrey SS, Thorsen T, Quist H, Matese JC, Brown PO, Botstein D, Eystein Lonning P, Borresen-Dale AL: Gene http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Breast Cancer Research Springer Journals

Prediction of breast cancer sensitivity to neoadjuvant chemotherapy based on status of DNA damage repair proteins

Loading next page...
 
/lp/springer-journals/prediction-of-breast-cancer-sensitivity-to-neoadjuvant-chemotherapy-k9KCzSlzWS

References (102)

Publisher
Springer Journals
Copyright
Copyright © 2010 by Asakawa et al.; licensee BioMed Central Ltd.
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology
eISSN
1465-542X
DOI
10.1186/bcr2486
pmid
20205718
Publisher site
See Article on Publisher Site

Abstract

Introduction: Various agents used in breast cancer chemotherapy provoke DNA double-strand breaks (DSBs). DSB repair competence determines the sensitivity of cells to these agents whereby aberrations in the repair machinery leads to apoptosis. Proteins required for this pathway can be detected as nuclear foci at sites of DNA damage when the pathway is intact. Here we investigate whether focus formation of repair proteins can predict chemosensitivity of breast cancer. Methods: Core needle biopsy specimens were obtained from sixty cases of primary breast cancer before and 18-24 hours after the first cycle of neoadjuvant epirubicin plus cyclophosphamide (EC) treatment. Nuclear focus formation of DNA damage repair proteins was immunohistochemically analyzed and compared with tumor response to chemotherapy. Results: EC treatment induced nuclear foci of gH2AX, conjugated ubiquitin, and Rad51 in a substantial amount of cases. In contrast, BRCA1 foci were observed before treatment in the majority of the cases and only decreased after EC in thirteen cases. The presence of BRCA1-, gH2AX-, or Rad51-foci before treatment or the presence of Rad51-foci after treatment was inversely correlated with tumor response to chemotherapy. DNA damage response (DDR) competence was further evaluated by considering all four repair indicators together. A high DDR score significantly correlated with low tumor response to EC and EC + docetaxel whereas other clinicopathological factors analyzed did not. Conclusions: High performing DDR focus formation resulted in tumor resistance to DNA damage-inducing chemotherapy. Our results suggested an importance of evaluation of DDR competence to predict breast cancer chemosensitivity, and merits further studying into its usefulness in exclusion of non-responder patients. Introduction tailoring treatment based on gene status significantly Recent advances in chemotherapy have significantly optimizes the response rate of hormone therapy and improved the prognosis of breast cancer patients. How- trastuzumab, respectively. Prediction of chemosensitivity ever, prediction of tumor sensitivity to chemotherapy with equivalent accuracy is currently anticipated to has not reached a high level of confidence, whereas further improve breast cancer prognosis. determining sensitivity to hormone therapy or trastuzu- Anthracycline-based regimens, such as epirubicin plus mab is relatively more established. Estrogen receptor cyclophosphamide (EC), and taxanes represent the (ER), progesterone receptor (PR) and human epidermal major chemotherapeutic agents used in the breast can- growth factor receptor (HER)2/ErbB2 are practical cer field [1,2]. Of these, anthracycline-based chemother- benchmarks to exclude non-responding patients, and apy induces DNA double-strand breaks (DSBs) [3,4], the most cytotoxic DNA lesion, that leads cells into apoptosis especially when relevant repair pathways * Correspondence: to@marianna-u.ac.jp 1 (represented by homologous recombination (HR) repair) Division of Breast and Endocrine Surgery, Department of Surgery, St. Marianna University School of Medicine, Kawasaki, 216-8511 Japan © 2010 Asakawa et al.; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 2 of 11 http://breast-cancer-research.com/content/12/2/R17 are perturbed [5]. It is important to note that DNA response to DNA damage-inducing chemotherapy and damage repair competence varies among individual whether it correlates with tumor fates after chemo- breast tumors and closely correlates with chemosensitiv- therapy, we analyzed foci in core needle biopsy speci- ity. For example, secondary mutations of BRCA1 or mens from breast cancer before and after neoadjuvant BRCA2 (essential factors in the HR pathway) caused by EC treatment. chemotherapy using cisplatin or poly(ADP-ribose) poly- merase inhibitor in BRCA1/2-mutated cancers restore Materials and methods the wild-type reading frame and, therefore, the tumor Patients and tumors acquires resistance to these drugs [6-8]. These facts Sixty patients with primary breast cancer (2 cm or lar- indicate that chemosensitivity of BRCA-associated can- ger) who consecutively underwent neoadjuvant che- cers could be strongly affected by DNA damage repair motherapy with EC followed by treatment with capability. Based on this evidence it has been suggested docetaxel (DOC) at the Division of Breast and Endo- that HR competence could be a potential biomarker for crine Surgery, St. Marianna University School of Medi- chemosensitivity [9]. Rad51, a protein that plays a direct cine, Japan, were enrolled in the present study from role in HR, especially reflects the HR competence of August 2005 to July 2007. Tumor specimens were cells. Therefore, knowing its status is likely to be valu- obtained by core needle biopsy prior to starting therapy able when assessing HR competence in tumor cells in and 18 to 24 hours after the first cycle of EC treatment. order to instruct therapeutic decisions [9]. Informed consent for the additional core needle biopsy The HR pathway for DSB repair is executed by and experimental use of tumor samples was obtained sequential recruitment of repair proteins to chromatin for all patients in accordance with an approved Institu- around DNA lesions. Accumulation of the proteins is tional Review Board application (registration number regulated by complex mechanisms that utilize phosphor- 946). ylation and ubiquitination modifications mediated by The chemotherapy regimen consisted of four 21-day 2 2 kinases including ataxia telangiectasia mutated (ATM), cycles of EC (E: 80 mg/m on day1,C:600 mg/m on and at least four ubiquitin E3 ligases, RNF8, RNF168, day 1) followed by four 21-day cycles of DOC (75 mg/ 2 2 Rad18, and BRCA1 [10-17]. The Mre11-Rad50-Nbs1 m on day 1). 75 mg/m DOC was administrated four complex first recognizes DSBs and recruits ATM. ATM times as total (only on day 1). There was no increase or then phosphorylates the histone variant H2AX (gH2AX) decrease of the dose. Tumor size was evaluated by [18,19] that triggers accumulation of the downstream E3 three-dimensional images obtained by helical computed ligases RNF8 [11-13,20] and RNF168 [14,15]. Lysine 63- tomography CT scan with a teleradiologic image work- linked polyubiquitin chains built at the sites of DNA station (ZIOSTATION®, Ziosoft Inc., Tokyo, Japan) at damage by these E3 ligases next recruits the BRCA1- baseline, 14 to 21 days after the last cycle of EC, and 21 Abraxas-RAP80 complex through the RAP80 compo- days after the last cycle of DOC treatment. The effect of nent, a protein that contains ubiquitin interacting motif chemotherapy on the tumor was assessed as the three- domains [21-23]. BRCA1 is then essential in order to dimensional volume reduction rate or tumor response rate. recruit repair effector proteins, including Rad51, that The tumor response was evaluated either by Response Eva- perform HR through sister chromatid exchange [24,25]. luation Criteria in Solid Tumors (RECIST) [26] or by the Depletionofany oneofthese proteins resultsinHR three-dimensional volume evaluation defined as: complete deficiency accompanied by loss of Rad51 focus forma- response (CR; disappearance of the disease), partial tion, causing cells to become hypersensitive to DSB- response (PR; reduction of tumor volume of ≥65%), stable inducing agents. disease (SD; volume reduction <65% or enlargement In this study we attempt to clarify the value of HR ≤73%), or progressive disease (PD; volume enlargement competence for the prediction of breast cancer chemo- ≥73%). These are equivalent to CR (disappearance), PR sensitivity. One contention is that nuclear focus forma- (reduction of ≥30%), SD (reduction <30% or enlargement tion of repair proteins in baseline breast cancer tissues ≤20%), or PD (enlargement ≥20%) in unidimensional is a response to spontaneous DNA damage during cell RECIST criteria, respectively (reviewed in [27]). We also proliferationand,inturn, mayrepresent amarkerof analyzed responses with a 50% border between PR and SD HR competence of cells to exogenous DNA damage. (instead of 65%) to evaluate more resistant cases. Therefore, it may predict tumor response to DNA damage-inducing chemotherapy such as with EC. Also, Immunohistochemical analysis the focus formation after chemotherapy could provide Immunohistochemical analysis was performed by us with additional information regarding the DNA the DAKO EnVision system (DAKO, Copenhagen, damage-response capacity. To verify in vivo whether Denmark) with modifications. Formalin-fixed, paraffin- focus formation of repair proteins actually occurs in embedded specimens were cut and heated in a water Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 3 of 11 http://breast-cancer-research.com/content/12/2/R17 bath (95°C, 40 minutes) in Target Retrieval Solution Statistical analysis (pH 9.0, Dako, Carpinteria, CA, USA) for detection of The variables measured in the study were first investi- BRCA-1 or in 10 mM sodium citrate buffer (pH 6.0) gated for association by the chi-squared contingency for gH2AX and Rad51. No pre-treatment was neces- table analysis. For rank correlation, Spearman’s method sary to detect conjugated ubiquitin. After quenching of was performed to determine the correlation between the foci score of two repair proteins and to determine the endogenous peroxidase, the sections were incubated correlation between tumor response rate and focus for- overnight at 4°C with primary antibody at the appro- mation of each repair protein or DDR score. For para- priate dilution [Additional file 1], washed with PBS, metric analyses of tumor volume reduction, Student’s and incubated with horseradish peroxidase-labeled polymer conjugated secondary antibody (EnVision+ unpaired t-test and the Tukey-Kramer method were System, Dako, Carpinteria, CA, USA) for 30 minutes at performed for two-factor comparisons and multiple room temperature. Color development was achieved by comparisons, respectively. For evaluation of significance 3, 3’-diaminobenzidine tetrahydrochloride. Effective- of DDR score and other clinicopathological factors in ness and specificity of each antibody for the detection correlation with mean tumor volume reduction or of DNA damage-induced nuclear foci were verified tumor response rate, variant analysis (univariate) or with cultured cells treated with ionizing radiation (IR) logistic regression analyses (univariate and multivaliate), or epirubicin. The immunofluorescent study has been respectively, were performed. All analyses were carried previously described [28,29]. The nuclear foci were out using Statview 5 statistical software (SAS Institute further analyzed with the protocol used in the tissue Inc, Cary, NC, USA). Statistical significance was stain. The intrinsic subtype[30] was approximated by declared for P values less than 0.05. receptor status determined by standard immunohisto- chemical and fluorescence in situ hybridization (FISH) Results analyses: luminal A: ER+ and/or PR+ and HER2-; Clinical and pathologic features luminal B: ER+ and/or PR+ and HER2+; HER2: ER- Sixty patients with primary breast cancer were included and PR- and HER2+; triple negative: ER- and PR- and in the present series. All tumors were diagnosed as inva- HER2-. Tumors that were immunochistochemically sive ductal carcinoma. Patient clinical characteristics are scored as 3+, or 2+ with FISH-positive, were regarded given in Table 1. All triple-negative tumors were posi- tive for CK5/6 (therefore described as basal-like in as positive for HER2 status. Cytokeratin (CK) 5/6 Table 1) whereas three cases of Luminal A, one case of expression was also examined to evaluate the basal-like Luminal B and three cases of HER2 type were positive character. for CK5/6. Three patients have one first-degree relative Immunohistochemical scoring with a history of breast cancer and two patients have Taking into consideration that all immunohistochem- one second-degree relative with a history of breast or ical markers used in the study localize to sites of DNA ovarian cancer. All patients completed an EC plus DOC damage in the normal HR pathway, we only counted cells displaying nuclear focus formation and disre- garded cytoplasmic or diffuse nuclear staining. We Table 1 Patient characteristics scored the nuclear foci staining as follows: 0 = no posi- Characteristic Number of Characteristic Number of tive cells, 1 = less than 10% positive cells, 2 = 10% or factor patients factor patients greater, but less than 80% positive cells, 3 = 80% or Age at treatment Cancer stage greater positive cells. Two observers (HA and HK) start were blinded to the clinical information to avoid Median 50 I 0 observer subjectivity when evaluating the immunohis- Range 34-68 II 53 tochemical staining. To correlate staining with tumor Lymph node metastasis III 5 response, we divided the cases into negative and posi- Negative 36 IV 2 tive samples to simplify the statistical analyses. The Positive 24 Intrinsic subtype* positive cases are a total of the categories with a foci Tumor stage Luminal A 37 score of 1, 2 and 3. To assess the capacity of the DNA T1 0 Luminal B 6 damage response (DDR) using a more comprehensive T2 54 HER2 11 approach, we configured the DDR score by counting T3 6 Basal-like 6 the total number of positive factors present in baseline T4 0 Total 60 foci of BRCA1, gH2AX and Rad51, and EC-induced * intrinsic subtypes were approximated by immunohistochemical receptor status. HER, human epidermal growth factor receptor. foci of Rad51, per case. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 4 of 11 http://breast-cancer-research.com/content/12/2/R17 regimen. Rad51 and gH2AX stains were not performed BRCA1 after EC (EC-induced foci score) significantly on tumor specimens before EC in two patients because correlated with that of Rad51 (P = 0.0017; Table 2), of insufficient tumor sample after reserving stocks for likely reflecting the requirement of BRCA1 for Rad51 clinical use. Tumor size evaluation by CT after treat- recruitment at the site of DNA damage. However, no ment with EC plus DOC was not performed for one other correlations between repair proteins were patient because of the patient’s condition. All but one observed, and no clear pattern combinations of repair patient received breast surgery after EC and DOC. proteins emerged. Nuclear foci staining of DNA damage repair proteins Association of focus formation of each repair protein with To assess the competence of the DSB repair pathway, tumor response to chemotherapy we immunohistochemically analyzed gH2AX, conjugated To elucidate the possible association between DDR com- ubiquitin, BRCA1, and Rad51 in nuclear foci based on petence and tumor response to chemotherapy, we corre- the idea that these candidates may represent a typical lated the presence of individual repair proteins in foci course of the DSB repair cascade [31]. Of these, gH2AX with tumor volume before and after chemotherapy. is the most upstream element, sequentially followed in Tumor volume was measured prior to chemotherapy to the cascade by conjugated ubiquitin, BRCA1, and establish the baseline volume. The mean volume reduc- Rad51. Rad51 is the most downstream of these four pro- tion of tumors after EC and after EC and DOC was 59.7 teins and is directly involved in HR. However, it should ± 25.8% and 76.0 ± 20.7% of baseline tumor volume, be mentioned that DNA repair failure due to genes at respectively. We analyzed the presence of repair proteins the same level of or downstream of RAD51, such as in foci before (baseline foci) and after EC treatment (EC- RAD51AP1 [32] or translesion DNA polymerases induced foci), sorted them into positive and negative foci [33,34], is an unlikely cause of loss of foci formation of groups for each individual repair protein, and then corre- these proteins. In addition to untreated, baseline breast lated each group with tumor volume [Additional file 2]. cancer tissues, we analyzed the tissues 18 to 24 hours There was a significant difference in tumor volume after after the first cycle of EC treatment to obtain further EC between BRCA1-positive and BRCA1-negative base- information for the assessment of DNA repair capacity. line foci groups (82.1 ± 17.8% vs 55.7 ± 25.1%, P = Theantibodiesusedinthisstudy arecommonlyused 0.0039) [Additional file 2a]. We then performed the same and well characterized in general. In addition we tested analysis after EC and DOC treatment. In addition to background staining and confirmed the specific detec- BRCA1 (93.7 ± 6.6% vs 72.8 ± 20.7%, P = 0.0044), signifi- tion of nuclear foci at DSBs caused by IR or epirubucin cant differences in tumor volume were observed between treatment (Figure 1). positive and negative gH2AX (78.4 ± 17.4% vs 65.6 ± The immunohistochemical analyzes revealed that in all 26.8%, P = 0.0429) and Rad51 baseline foci groups (78.1 but two cases, the foci score of at least one of the repair ± 18.9% vs 63.6 ± 24.4%, P = 0.0351) [Additional file 2b]. proteins was altered in response to EC treatment. We next tested the correlation between scored foci Representative data for immunohistochemical findings groups and the tumor response rate. The tumor of the nuclear focus formation of the repair proteins response rate was evaluated with RECIST or three- before and after the first cycle of EC are shown in dimensional volume reduction using either 65% or 50% Figure 2 with panels summarizing the foci scores of the of the PR/SD border (as described in the Materials and cases. Prior to EC treatment, samples were stained to Methods). Tumor responses to EC and EC plus DOC determine baseline staining of foci. The foci were posi- according to focus formation status are shown in Addi- tive for gH2AX (20 of58cases), BRCA1(51 of 60 tional file 3. Contingency table analyses demonstrate sig- cases), or Rad51 (11 of 58 cases) whereas no cases nificant differences in the EC tumor response rate exhibited foci staining for conjugated ubiquitin (0 of 60 between BRCA1-positive and BRCA1-negative baseline cases). In response to EC treatment, the number of foci foci groups and between Rad51-positive and Rad51- staining positive for gH2AX (44 of 58 cases), conjugated negative EC-induced foci groups for all three criteria of ubiquitin (26 of 60 cases), and Rad51 (31 of 58 cases) the response rate (ZIO 65%, ZIO 50%, RECIST). There increased, whereas foci staining for BRCA1 either continued to be a significant difference in tumor increased (9 of 60 cases), remained unchanged (38 of 60 response rate after EC and DOC treatment between cases) or decreased (13 of 60 cases). The reason why Rad51-positive and Rad51-negative EC-induced foci BRCA1 foci staining decreased after treatment in some groups for all three criteria. In addition, when evaluated cases is not clear at present but it could be implicated with a three-dimensional volume reduction using 50% of in the presence of BRCA1 foci in normal S-phase that the PR/SD border, significant differences in the tumor colocalizes with proliferating cell nuclear antigen response rate to EC and DOC were observed between (PCNA) at DNA replication fork [35]. The foci score of Rad51-positive and Rad51-negative baseline foci groups. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 5 of 11 http://breast-cancer-research.com/content/12/2/R17 Figure 1 Immunohistochemistry controls and antibody specificity. (a) Immunohistochemical staining with control IgG for tumors after the first cycle of epirubicin plus cyclophosphamide (EC) treatment (upper panel). Lower panel shows Rad51 staining for morphologically diagnosed non-cancerous breast tissues (left upper part) and tumor (right lower part) after the first cycle of EC treatment. Although non-cancerous breast cells also expressed nuclear foci formation the number and intensity was significantly lower than that in tumor cells. (b and c) DNA damage- induced nuclear foci formations detected by antibodies used in the study. HeLa cells were either untreated (-), treated with 5 Gy ionizing radiation (IR) or 0.2 μg/ml epirubicin (Epi), incubated for three hours and fixed. Cells were then subjected either to immunofluorescent analyses with the indicated primary antibodies and (b) FITC- (green) or Rhodamine- (red) conjugated secondary antibodies, or (c) to the same protocol as that used in the tissue stain. For immunofluorescent analyses the nucleus was counterstained with DAPI. Ub, conjugated ubiquitin. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 6 of 11 http://breast-cancer-research.com/content/12/2/R17 Figure 2 Nuclear focus formation in response to chemotherapy. Tumor specimens were obtained by core needle biopsy before and 18 to 24 hours after the first cycle of epirubicin plus cyclophosphamide (EC) treatment. Immunohistochemical findings from representative cases for gH2AX, conjugated ubiquitin (Ub), BRCA1, and Rad51 are shown. Graphs at right demonstrate changes in nuclear foci score after EC treatment in all cases analyzed. The red, blue, and black lines indicate cases with increased, decreased, and unchanged scores, respectively. The thickness of the lines proportionally reflects the number of cases. The thinnest line (gH2AX score 3 to 3) and the thickest line (Ub, score 0 to 0) represent 1 and 34 cases, respectively. n, number of cases analyzed. To specify the correlation of these focus formation Table 2 Correlation between EC-induced foci of Rad51 and BRCA1 groups with tumor response rates, we further analyzed thedatawithSpearman ’s rank correlation method. BRCA1 When evaluated with three-dimensional volume reduc- 0 1 2 3 Total tion using 50% of the PR/SD border, Spearman’sanaly- Rad51 sis showed that the presence of BRCA1-positive baseline 0 7 10 2 0 19 foci associated with poor EC tumor response (P = 1 3 24 9 2 38 0.0067) [Additional file 3]. Spearman’sanalysisalso 20 0 2 0 2 demonstrated that the presence of Rad51-positive base- Total 10 34 13 2 59 line foci (P = 0.0078) or EC-induced foci (P = 0.0042) P = 0.0017 [Additional file 3] associated with poor EC and DOC P value is from the Spearman’s rank correlation test. EC, epirubicin plus tumor response. cyclophosphamide. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 7 of 11 http://breast-cancer-research.com/content/12/2/R17 Association of DDR score with tumor reduction by chemotherapy The analysis correlating focus formation of BRCA1, gH2AX, and Rad51 prior to treatment and of Rad51 foci after EC treatment with the mean tumor volume reduc- tion or tumor response rate [Additional files 2 and 3] uncovers a significant inverse correlation with tumor response for each of the four conditions. These data sup- port the supposition that higher DDR competency pro- duces tumors resistant to chemotherapy. To correlate overall DDR competency with tumor reduction, we devised a simple measurement to assess DDR competency. Each patient case was analyzed for the presence of all four of the above listed conditions and was assigned a DDR score of 0 to 4 based on the number of conditions present. This DDR score was then correlated with mean tumor volume reductions. Number of cases in each DDR score is shown in Table 3. As shown in Figure 3, both the mean tumor volume reductions after EC (28.4 ± 28.1%) and afterECand DOC (49.9 ±22.0%) for DDR score4(all four conditions present) were the lowest among all the scores. There were significant differences between score 4 and either score 0 or 2 for the mean tumor volume reduc- tions after EC (Figure 3a) and between score 4 and either score 0, 1 or 2 for the mean tumor volume reductions after EC and DOC (Figure 3b), as judged by the Tukey- Kramer multiple comparisons study setting P <0.05asa significance threshold. In addition, Spearman’sanalysis showed that a high DDR score was associated with poor tumor response rate after EC and DOC (P = 0.0031) when evaluated with three-dimensional volume reduction using 50% of the PR/SD border (Table 3). A high DDR score also tended to be associated with poor tumor response rate after EC (P = 0.0639, Table 3). Table 3 Correlation between DDR score and tumor response rate (ZIO 50%) Figure 3 Mean tumor volume reductions after (a) EC or (b) EC DDR score and DOC according to DDR score. Error bars represent standard 0 1 2 3 4 Total deviation. Significance was analyzed by Tukey-Kramer test setting P < 0.05 as significance threshold. DDR, DNA damage response; DOC, After EC docetaxel; EC, epirubicin plus cyclophosphamide. CR 1 0 1 0 0 2 PR 1 12 14 9 2 38 SD 0 3 8 3 4 18 The correlation between DDR score and the tumor Total 2 15 23 12 6 58 response prompted us to examine whether it has a sig- P = 0.0639 nificant impact among other clinicopathological factors After EC + DOC including age, cancer stage, tumor size, nodal metastatic CR 1 1 3 1 0 6 status, and subtypes. None of these factors correlated PR 1 13 19 9 2 44 with DDR score (data not shown). The variant analysis SD 0 0 1 2 4 7 for mean tumor volume reduction after EC revealed Total 2 14 23 12 6 57 that only DDR score (P = 0.0069), but no other factors P = 0.0031 correlated with the mean tumor volume reduction P values are from the Spearman’s rank correlation test. CR, complete response; (Table 4). The variant analysis for mean tumor volume DDR, DNA damage response; EC, epirubicin plus cyclophosphamide; PR, partial response; SD, stable disease. reduction after EC and DOC also demonstrated DDR Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 8 of 11 http://breast-cancer-research.com/content/12/2/R17 score (P = 0.0035) as the most significant correlation patients and we also considered the patient’sconveni- factor, followed by nodal status (P = 0.0201) and tumor ence. However, the ideal timing remains to be deter- size (P = 0.0538, Table 4). In addition, univariate logistic mined if biopsy after chemotherapy is required. regression analysis showed that a high (3 and 4) DDR The second reason for the diversity of the DDR score was most significantly associated with poor tumor response could be attributed to the diversity of aberra- response after EC and DOC (P = 0.0095) when evalu- tions of the genes responsible for DSB repair in each ated with volume reductionusing 50%ofthe PR/SD breast cancer. Theoretically, defects in the recruitment border, followed by tumor size (P = 0.0260), cancer of upstream repair proteins could result in loss of down- stage (P = 0.0465), and subtype (P = 0.0659, Table 5). stream proteins at sites of DNA damage, and this has We then examined multivariate analysis with tumor been shown to be the case in many molecular biological size, nodal status, subtype, and DDR score, the factors studies using cultured cells [10-15,21-23]. Furthermore, that showed probable association with tumor response it was also shown that Rad51 nuclear expression is rate in the univariate analysis. Cancer stage was omitted absent in tumors associated with BRCA2 mutation [41]. because it was correlated with tumor size. Importantly Thepositivecorrelation foundbetween EC-induced the result indicated that only the DDR score was signifi- BRCA1 and Rad51 foci in this study (Table 2) may also cantly associated with tumor response rate (P =0.0402) support this interpretation. In contrast, it was reported independent of other factors analyzed (Table 6). that overexpression of Rad51 restored Rad51 focus for- mation and rescued the sensitivity of BRCA1-deficient Discussion cells to x-rays and cisplatin [42]. Importantly, up-regula- In the present study using human tumor specimens we tion of Rad51 was a common feature of BRCA1-defi- show for the first time that DNA repair competence may cient breast tumors [42]. These data suggest that the predict breast cancer sensitivity to DNA damage-inducing mechanism of DSB repair response in vivo is not simple chemotherapy. We selected gH2AX, conjugated ubiquitin, and that assessment of DSB repair aberrations in each BRCA1, and Rad51, proteins in the DSB repair cascade, to patient case is, therefore, unreasonable at present. assess DNA repair competence because accumulated evi- In an attempt to address this problem in our current dence demonstrates that inactivation of genes in the DSB study, we assessed the comprehensive capacity of DSB repair pathway results in cellular sensitivity to DNA repair by incorporating multiple candidate factors into damage-inducing chemotherapy [16,29,31,36-38]. In our one DDR score. We found that foci of BRCA1, gH2AX, study, these repair proteins dramatically responded to EC and Rad51 prior to treatment and EC-induced foci of treatment. The conjugated ubiquitin response was espe- Rad51 correlated with tumor response when compared cially dramatic as approximately half of the cases analyzed either with the mean tumor volume reduction or the formed conjugated ubiquitin foci, compared with unde- tumor response rate. When incorporating these four fac- tectable foci formation prior to treatment. This suggests tors into one DDR score a significant correlation was that ubiquitination occurs in vivo during the DNA damage observed with mean tumor volume reduction after EC, response in an early stage after chemotherapy. However, whereas no other factors correlated with the mean in spite of the dramatic response, we did not find any sig- tumor volume reduction (Table 4 and Figure 3a). nificant correlation between conjugated ubiquitin foci for- Although it was not statistically significant, the similar mation and tumor response. The reason is currently correlation was also observed between DDR score and unknown. One possibility is that this could be attributed tumor response rate (Table 3). These correlations to the fact that ubiquitination is also involved in DNA became more significant after EC and DOC treatment damage response pathways other than for DSBs. (Tables 3 to 5 and Figure 3b) and the DDR score was We did not find certain trends of the combinations of an independent predictive factor of other factors includ- responding repair proteins. Several reasons could ing tumor subtype when evaluated with volume reduc- account for this observation. First, the metabolism and tion using 50% of the PR/SD border (Table 6). Recent pharmacokinetics of the agents could vary per patient. studies suggested that luminal tumors have low response The ideal time to obtain the in vivo sample was, there- rate to neoadjuvant chemotherapy, whereas basal-like fore, difficult to determine. The experimental design and HER2+ tumors have higher response rates. For employed in this study was not very robust in this way. example, it has been reported that clinical response rate In cultured cells, gH2AX accumulates at sites of DNA (CR and PR) to anthracyclin-based chemotherapy of damage just minutes after the damage occurs, whereas luminal A was 39%, whereas that of basal-like, which BRCA1 and Rad51 foci appear 30 minutes to several has been implicated with BRCA1 dysfunction [43,44], hours afterwards [11,35,39,40]. In this study we har- was 85% [45]. The response rates to EC treatment of vested samples 18 to 24 hours after EC treatment luminal A (15 of 37 cases, 40.5%) and basal-like (4 of 6 because the agents were still expected to be present in cases, 66.7%) subtypes in the current study were not Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 9 of 11 http://breast-cancer-research.com/content/12/2/R17 Table 4 Univariate analysis of variance for mean tumor Table 4: Univariate analysis of variance for mean tumor volume reduction volume reduction (Continued) n Mean tumor volume P 3 12 70.7 ± 27.5 reduction (%) ± SD 4 6 49.9 ± 22.0 After EC DDR, DNA Damage Response; DOC, docetaxel; EC, epirubicin plus Age (years) cyclophosphamide; HER, human epidermal growth factor receptor; SD, standard deviation. -50 30 64.1 ± 25.9 0.1710 51- 29 54.9 ± 25.4 Cancer stage Table 5 Univariate logistic regression analysis of factors II 53 61.5 ± 25.1 0.0962 affecting tumor response rate (ZIO 50%) III 5 55.7 ± 29.1 Odds ratio (95% CI) P IV 2 21.9 ± 6.4 After EC + DOC Tumor stage Age (years) T2 54 60.8 ± 25.3 0.3059 <51 1.000 T3 6 21.9 ± 6.4 51≤ 0.800 (0.192-3.333) 0.7592 Nodal status Cancer stage N- 24 62.7 ± 24.9 0.4557 II 1.000 N+ 36 57.6 ± 26.6 III, IV 5.750 (1.028-32.174) 0.0465 Subtype Tumor stage Luminal A 37 58.8 ± 21.3 0.2923 T2 1.000 Luminal B 6 72.9 ± 9.4 T3 7.833 (1.279-47.964) 0.0260 HER2 11 50.4 ± 36.4 Nodal status Basal-like 6 69.4 ± 37.0 N- 1.000 DDR score N+ 6.286 (0.730-54.110) 0.0942 0 2 94.6 ± 7.6 0.0069 Subtype 1 15 57.3 ± 19.4 Luminal A, B 1.000 2 23 62.8 ± 22.9 HER2, Basal-like 3.958 (0.913-17.154) 0.0659 3 12 61.1 ± 26.6 DDR score 4 6 28.4 ± 28.1 0, 1, 2 1.000 After EC-DOC 3, 4 9.423 (1.729-51.359) 0.0095 Age CI, confidence interval; DDR, DNA Damage Response; DOC, docetaxel; EC, -50 30 80.6 ± 23.9 0.0804 epirubicin plus cyclophosphamide; HER, human epidermal growth factor receptor. 51- 29 71.2 ± 15.7 Cancer stage very different from the previous report. However, we II 52 77.9 ± 19.8 0.1230 could not find any correlation between subtype and III 5 64.3 ± 25.7 DDR score while DDR score independently predicted IV 2 54.7 ± 17.6 the chemosensitivity. The result may reflect the fact that Tumor stage luminal A tumors also include DNA damage-sensitive T2 53 77.7 ± 19.7 0.0538 tumors with defective HR pathways that can be counted T3 6 60.6 ± 24.7 by the DDR score. Supporting this it has been shown Nodal status that tumors caused by BRCA2 deficiency mainly become N- 23 83.7 ± 14.3 0.0201 luminal A tumors [44,46,47]. N+ 36 71.0 ± 22.7 The reason why the correlation between the DDR score Subtype and tumor response after EC and DOC treatment became Luminal A 36 77.9 ± 16.1 0.0789 more significant than that after EC is not clear at present. Luminal B 6 86.0 ± 9.5 As DOC does not induce DNA DSBs, the observed effect HER2 11 62.4 ± 30.5 is not likely to be due to the sensitivity to DNA damage in Basal-like 6 79.7 ± 24.8 those tumors. DOC might be more toxic for the cells with DDR score gross genomic aberration caused by the pretreatment with 0 2 99.6 ± 0.6 0.0035 EC under the condition of being less HR competent. 1 14 77.2 ± 13.2 Alternatively it is possible that time length after EC treat- 2 23 81.0 ± 14.7 ment enhanced the difference of the outcome. Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 10 of 11 http://breast-cancer-research.com/content/12/2/R17 epirubicin plus cyclophosphamide; ER: estrogen receptor; HER: human Table 6 Multivariate logistic regression analysis of factors epidermal growth factor receptor; HR: homologous recombination repair; IR: affecting tumor response rate (ZIO 50%) ionizing radiation; PBS: phosphate-buffered saline; PD: progressive disease; Odds ratio (95% CI) P PR: partial response; PR (+ or -): progesterone receptor (+ or -); RECIST: Response Evaluation Criteria in Solid Tumors; SD: stable disease. After EC + DOC Tumor stage Acknowledgements T2 1.000 This study was supported by grants from the Ministry of Education, Science, Sports, Culture and Technology of Japan grants-in-aid. We thank Mrs. T3 2.246 (0.290-17.420) 0.4388 Shigeko Ohnuma for technical support for immunohistochemical Nodal status experiments; Dr. Yasuo Miyoshi for technical advice; Dr. Ettore Appella for N- 1.000 critical reading of the manuscript; Members in the Division of Breast and Endocrine Surgery for support in collection of biopsy specimens; and Mrs. N+ 3.651 (0.346-38.506) 0.2813 Rie Ogawa and Miss Umi Noguchi for secretarial assistance. Subtype Luminal A, B 1.000 Author details Division of Breast and Endocrine Surgery, Department of Surgery, St. HER2, Basal-like 2.484 (0.464-13.287) 0.2874 Marianna University School of Medicine, Kawasaki, 216-8511 Japan. DDR score Department of Translational Oncology, St. Marianna University Graduate 0, 1, 2 1.000 School of Medicine, Kawasaki, 216-8511 Japan. Department of Diagnostic Pathology, St. Marianna University School of Medicine, Kawasaki, 216-8511 3, 4 6.694 (1.088-41.182) 0.0402 Japan. Department of Breast and Endocrine Surgery, Kumamoto University, CI, confidence interval; DDR, DNA Damage Response; DOC, docetaxel; EC, Honjo 1-1-1, Kumamoto 860-8556, Japan. epirubicin plus cyclophosphamide; HER, human epidermal growth factor receptor. Authors’ contributions HA analyzed the majority of the data. HK conducted immunohistochemical analyses. AK obtained the data for tumor response. MT supported Interestingly, DDR score group 4 consisted of cases immunohistochemical analyses. WW characterized antibody specificities. HI with poor tumor responses to chemotherapy when eval- and MF made substantial contributions to analysis and interpretation of the data. TO designed and conducted the studies, and wrote the manuscript. All uated for both mean tumor volume reduction (Figure 3) authors read and approved the final manuscript. and tumor response rate (Table 3). This result may lead to the possibility of using DDR status in the clinic to Competing interests The authors declare that they have no competing interests. predict and exclude non-responders to EC treatment. It is noteworthy to point out that the HR repair cascade Received: 18 September 2009 Revised: 7 January 2010 for DSB contains many essential proteins other than Accepted: 5 March 2010 Published: 5 March 2010 those tested in this study. By including select subsets of References proteins for analysis, it may be possible to identify non- 1. Trudeau M, Charbonneau F, Gelmon K, Laing K, Latreille J, Mackey J, responders in order to avoid unnecessary chemotherapy. McLeod D, Pritchard K, Provencher L, Verma S: Selection of adjuvant Ideally in such cases, the levels of baseline foci present chemotherapy for treatment of node-positive breast cancer. Lancet Oncol 2005, 6:886-898. prior to treatment would provide enough information to 2. Effects of chemotherapy and hormonal therapy for early breast cancer determine appropriate treatment, preventing the need on recurrence and 15-year survival: an overview of the randomised for additional core needle biopsy after chemotherapy. trials. Lancet 2005, 365:1687-1717. 3. Capranico G, Zunino F, Kohn KW, Pommier Y: Sequence-selective topoisomerase II inhibition by anthracycline derivatives in SV40 DNA: Conclusions relationship with DNA binding affinity and cytotoxicity. Biochemistry 1990, In conclusion, our results suggest the importance of 29:562-569. 4. Capranico G, De Isabella P, Penco S, Tinelli S, Zunino F: Role of DNA evaluating DDR competence to predict breast cancer breakage in cytotoxicity of doxorubicin, 9-deoxydoxorubicin, and chemosensitivity and warrant further investigation into 4-demethyl-6-deoxydoxorubicin in murine leukemia P388 cells. Cancer itseffectivenessasawaytoexclude non-responding Res 1989, 49:2022-2027. 5. Sancar A, Lindsey-Boltz LA, Unsal-Kacmaz K, Linn S: Molecular mechanisms patients. of mammalian DNA repair and the DNA damage checkpoints. Annu Rev Biochem 2004, 73:39-85. Additional file 1: Table S1. Antibodies used in the present 6. Sakai W, Swisher EM, Karlan BY, Agarwal MK, Higgins J, Friedman C, immunohistochemical study. Villegas E, Jacquemont C, Farrugia DJ, Couch FJ, Urban N, Taniguchi T: Secondary mutations as a mechanism of cisplatin resistance in BRCA2- Additional file 2: Figure S1. Mean tumor volume reduction after EC (a) mutated cancers. Nature 2008, 451:1116-1120. or EC+DOC (b) according to the nuclear foci status for DNA repair 7. Edwards SL, Brough R, Lord CJ, Natrajan R, Vatcheva R, Levine DA, Boyd J, proteins. Reis-Filho JS, Ashworth A: Resistance to therapy caused by intragenic Additional file 3: Table S2. Tumor response rate according to the deletion in BRCA2. Nature 2008, 451:1111-1115. nuclear foci status for DNA repair proteins. 8. Swisher EM, Sakai W, Karlan BY, Wurz K, Urban N, Taniguchi T: Secondary BRCA1 mutations in BRCA1-mutated ovarian carcinomas with platinum resistance. Cancer Res 2008, 68:2581-2586. 9. Livingston DM, Silver DP: Cancer: crossing over to drug resistance. Nature Abbreviations 2008, 451:1066-1067. CK: cytokeratin; CR: complete response; CT: computed tomography; DDR: DNA damage response; DOC: docetaxel; DSB: DNA double-strand break; EC: Asakawa et al. Breast Cancer Research 2010, 12:R17 Page 11 of 11 http://breast-cancer-research.com/content/12/2/R17 10. Wang B, Elledge SJ: Ubc13/Rnf8 ubiquitin ligases control foci formation expression patterns of breast carcinomas distinguish tumor subclasses of the Rap80/Abraxas/Brca1/Brcc36 complex in response to DNA with clinical implications. Proc Natl Acad Sci USA 2001, 98:10869-10874. damage. Proc Natl Acad Sci USA 2007, 104:20759-20763. 31. Harper JW, Elledge SJ: The DNA damage response: ten years after. Mol 11. Mailand N, Bekker-Jensen S, Faustrup H, Melander F, Bartek J, Lukas C, Cell 2007, 28:739-745. Lukas J: RNF8 Ubiquitylates Histones at DNA Double-Strand Breaks and 32. Wiese C, Dray E, Groesser T, San Filippo J, Shi I, Collins DW, Tsai MS, Promotes Assembly of Repair Proteins. Cell 2007, 131:887-900. Williams GJ, Rydberg B, Sung P, Schild D: Promotion of homologous 12. Kolas NK, Chapman JR, Nakada S, Ylanko J, Chahwan R, Sweeney FD, recombination and genomic stability by RAD51AP1 via RAD51 Panier S, Mendez M, Wildenhain J, Thomson TM, Pelletier L, Jackson SP, recombinase enhancement. Mol Cell 2007, 28:482-490. Durocher D: Orchestration of the DNA-damage response by the RNF8 33. Kawamoto T, Araki K, Sonoda E, Yamashita YM, Harada K, Kikuchi K, ubiquitin ligase. Science 2007, 318:1637-1640. Masutani C, Hanaoka F, Nozaki K, Hashimoto N, Takeda S: Dual roles for 13. Huen MS, Grant R, Manke I, Minn K, Yu X, Yaffe MB, Chen J: RNF8 DNA polymerase eta in homologous DNA recombination and translesion Transduces the DNA-Damage Signal via Histone Ubiquitylation and DNA synthesis. Mol Cell 2005, 20:793-799. Checkpoint Protein Assembly. Cell 2007, 131:901-914. 34. McIlwraith MJ, Vaisman A, Liu Y, Fanning E, Woodgate R, West SC: Human 14. Doil C, Mailand N, Bekker-Jensen S, Menard P, Larsen DH, Pepperkok R, DNA polymerase eta promotes DNA synthesis from strand invasion Ellenberg J, Panier S, Durocher D, Bartek J, Lukas J, Lukas C: RNF168 binds intermediates of homologous recombination. Mol Cell 2005, 20:783-792. and amplifies ubiquitin conjugates on damaged chromosomes to allow 35. Scully R, Chen J, Ochs RL, Keegan K, Hoekstra M, Feunteun J, accumulation of repair proteins. Cell 2009, 136:435-446. Livingston DM: Dynamic changes of BRCA1 subnuclear location and 15. Stewart GS, Panier S, Townsend K, Al-Hakim AK, Kolas NK, Miller ES, phosphorylation state are initiated by DNA damage. Cell 1997, Nakada S, Ylanko J, Olivarius S, Mendez M, Oldreive C, Wildenhain J, 90:425-435. Tagliaferro A, Pelletier L, Taubenheim N, Durandy A, Byrd PJ, Stankovic T, 36. Murakawa Y, Sonoda E, Barber LJ, Zeng W, Yokomori K, Kimura H, Niimi A, Taylor AM, Durocher D: The RIDDLE syndrome protein mediates a Lehmann A, Zhao GY, Hochegger H, Boulton SJ, Takeda S: Inhibitors of the ubiquitin-dependent signaling cascade at sites of DNA damage. Cell proteasome suppress homologous DNA recombination in mammalian 2009, 136:420-434. cells. Cancer Res 2007, 67:8536-8543. 16. Wu W, Koike A, Takeshita T, Ohta T: The ubiquitin E3 ligase activity of 37. Collis SJ, Swartz MJ, Nelson WG, DeWeese TL: Enhanced radiation and BRCA1 and its biological functions. Cell Div 2008, 3:1. chemotherapy-mediated cell killing of human cancer cells by small 17. Huang J, Huen MS, Kim H, Leung CC, Glover JN, Yu X, Chen J: RAD18 inhibitory RNA silencing of DNA repair factors. Cancer Res 2003, transmits DNA damage signalling to elicit homologous recombination 63:1550-1554. repair. Nat Cell Biol 2009, 11:592-603. 38. Foulkes WD: BRCA1 and BRCA2: chemosensitivity, treatment outcomes 18. Bassing CH, Suh H, Ferguson DO, Chua KF, Manis J, Eckersdorff M, and prognosis. Fam Cancer 2006, 5:135-142. Gleason M, Bronson R, Lee C, Alt FW: Histone H2AX: a dosage-dependent 39. Ayoub N, Jeyasekharan AD, Bernal JA, Venkitaraman AR: HP1-beta suppressor of oncogenic translocations and tumors. Cell 2003, mobilization promotes chromatin changes that initiate the DNA damage 114:359-370. response. Nature 2008, 453:682-686. 19. Celeste A, Fernandez-Capetillo O, Kruhlak MJ, Pilch DR, Staudt DW, Lee A, 40. Zhong Q, Chen CF, Li S, Chen Y, Wang CC, Xiao J, Chen PL, Sharp ZD, Bonner RF, Bonner WM, Nussenzweig A: Histone H2AX phosphorylation is Lee WH: Association of BRCA1 with the hRad50-hMre11-p95 complex dispensable for the initial recognition of DNA breaks. Nat Cell Biol 2003, and the DNA damage response. Science 1999, 285:747-750. 5:675-679. 41. Honrado E, Osorio A, Palacios J, Milne RL, Sanchez L, Diez O, Cazorla A, 20. Plans V, Scheper J, Soler M, Loukili N, Okano Y, Thomson TM: The RING Syrjakoski K, Huntsman D, Heikkila P, Lerma E, Kallioniemi A, Rivas C, finger protein RNF8 recruits UBC13 for lysine 63-based self Foulkes WD, Nevanlinna H, Benitez J: Immunohistochemical expression of polyubiquitylation. J Cell Biochem 2006, 97:572-582. DNA repair proteins in familial breast cancer differentiate BRCA2- 21. Kim H, Chen J, Yu X: Ubiquitin-binding protein RAP80 mediates BRCA1- associated tumors. J Clin Oncol 2005, 23:7503-7511. dependent DNA damage response. Science 2007, 316:1202-1205. 42. Martin RW, Orelli BJ, Yamazoe M, Minn AJ, Takeda S, Bishop DK: RAD51 up- 22. Sobhian B, Shao G, Lilli DR, Culhane AC, Moreau LA, Xia B, Livingston DM, regulation bypasses BRCA1 function and is a common feature of BRCA1- Greenberg RA: RAP80 targets BRCA1 to specific ubiquitin structures at deficient breast tumors. Cancer Res 2007, 67:9658-9665. DNA damage sites. Science 2007, 316:1198-1202. 43. Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, Deng S, 23. Wang B, Matsuoka S, Ballif BA, Zhang D, Smogorzewska A, Gygi SP, Johnsen H, Pesich R, Geisler S, Demeter J, Perou CM, Lonning PE, Elledge SJ: Abraxas and RAP80 form a BRCA1 protein complex required Brown PO, Borresen-Dale AL, Botstein D: Repeated observation of breast for the DNA damage response. Science 2007, 316:1194-1198. tumor subtypes in independent gene expression data sets. Proc Natl 24. Greenberg RA, Sobhian B, Pathania S, Cantor SB, Nakatani Y, Livingston DM: Acad Sci USA 2003, 100:8418-8423. Multifactorial contributions to an acute DNA damage response by 44. Turner N, Tutt A, Ashworth A: Hallmarks of ‘BRCAness’ in sporadic BRCA1/BARD1-containing complexes. Genes Dev 2006, 20:34-46. cancers. Nat Rev Cancer 2004, 4:814-819. 25. Yu DS, Sonoda E, Takeda S, Huang CL, Pellegrini L, Blundell TL, 45. Carey LA, Dees EC, Sawyer L, Gatti L, Moore DT, Collichio F, Ollila DW, Venkitaraman AR: Dynamic control of Rad51 recombinase by self- Sartor CI, Graham ML, Perou CM: The triple negative paradox: primary association and interaction with BRCA2. Mol Cell 2003, 12:1029-1041. tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res 2007, 26. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, 13:2329-2334. Verweij J, Van Glabbeke M, van Oosterom AT, Christian MC, Gwyther SG: 46. Lakhani SR, Vijver Van De MJ, Jacquemier J, Anderson TJ, Osin PP, New guidelines to evaluate the response to treatment in solid tumors. McGuffog L, Easton DF: The pathology of familial breast cancer: European Organization for Research and Treatment of Cancer, National predictive value of immunohistochemical markers estrogen receptor, Cancer Institute of the United States, National Cancer Institute of progesterone receptor, HER-2, and p53 in patients with mutations in Canada. J Natl Cancer Inst 2000, 92:205-216. BRCA1 and BRCA2. J Clin Oncol 2002, 20:2310-2318. 27. Jaffe CC: Measures of response: RECIST, WHO, and new alternatives. J 47. Foulkes WD, Metcalfe K, Sun P, Hanna WM, Lynch HT, Ghadirian P, Tung N, Clin Oncol 2006, 24:3245-3251. Olopade OI, Weber BL, McLennan J, Olivotto IA, Begin LR, Narod SA: 28. Sato K, Hayami R, Wu W, Nishikawa T, Nishikawa H, Okuda Y, Ogata H, Estrogen receptor status in BRCA1- and BRCA2-related breast cancer: Fukuda M, Ohta T: Nucleophosmin/B23 is a candidate substrate for the the influence of age, grade, and histological type. Clin Cancer Res 2004, BRCA1-BARD1 ubiquitin ligase. J Biol Chem 2004, 279:30919-30922. 10:2029-2034. 29. Takeshita T, Wu W, Koike A, Fukuda M, Ohta T: Perturbation of DNA repair doi:10.1186/bcr2486 pathways by proteasome inhibitors corresponds to enhanced Cite this article as: Asakawa et al.: Prediction of breast cancer sensitivity chemosensitivity of cells to DNA damage-inducing agents. Cancer to neoadjuvant chemotherapy based on status of DNA damage repair Chemother Pharmacol 2009, 64:1039-46. proteins. Breast Cancer Research 2010 12:R17. 30. Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H, Hastie T, Eisen MB, Rijn van de M, Jeffrey SS, Thorsen T, Quist H, Matese JC, Brown PO, Botstein D, Eystein Lonning P, Borresen-Dale AL: Gene

Journal

Breast Cancer ResearchSpringer Journals

Published: Mar 5, 2010

There are no references for this article.