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

Learn More →

MRI for nodal restaging after neoadjuvant therapy in rectal cancer with histopathologic comparison

MRI for nodal restaging after neoadjuvant therapy in rectal cancer with histopathologic comparison Background After neoadjuvant therapy, most of the lymph nodes (LNs) will shrink and disappear in patients with rectal cancer. However, LNs that are still detectable on MRI carry a risk of metastasis. This study aimed to evaluate the performance of the European Society of Gastrointestinal and Abdominal Radiology (ESGAR) criterion (short-axis diameter ≥ 5 mm) in diagnosing malignant LNs in patients with rectal cancer after neoadjuvant therapy, and whether nodal morphological characteristics (including shape, border, signal homogeneity, and enhancement homogeneity) could improve the diagnostic efficiency for LNs ≥ 5 mm. Methods This retrospective study included 90 patients with locally advanced rectal cancer who underwent surgery after neoadjuvant therapy and performed preoperative MRI. Two radiologists independently measured the short- axis diameter of LNs and evaluated the morphological characteristics of LNs ≥ 5 mm in consensus. With a per node comparison with histopathology as the reference standard, a ROC curve was performed to evaluate the diagnostic performance of the size criterion. For categorical variables, either a χ test or Fisher’s exact test was used. Results A total of 298 LNs were evaluated. The AUC for nodal size in determining nodal status was 0.81. With a size cutoff value of 5 mm, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 65.9%, 87.0%, 46.8%, 93.6% and 83.9%, respectively. No significant differences were observed in any of the morphological characteristics between benign and malignant LNs ≥ 5 mm (all P > 0.05). Conclusions The ESGAR criterion demonstrated moderate diagnostic performance in identifying malignant LNs in patients with rectal cancer after neoadjuvant therapy. It was effective in determining the status of LNs < 5 mm but not for LNs ≥ 5 mm, and the diagnostic efficiency could not be improved by considering nodal morphological characteristics. Keywords Rectal cancer, Neoadjuvant therapy, Lymph node, Magnetic resonance imaging, Histopathology *Correspondence: Shenping Yu yushp@mail.sysu.edu.cn Department of Radiology, The First Affiliated Hospital, Sun Yat-sen University, 510080 Guangzhou, China Department of Radiology, The Seventh Affiliated Hospital, Sun Yat-sen University, 518036 Shenzhen, China MR Scientific Marketing, SIEMENS Healthineers Ltd, 200124 Shanghai, China © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhang et al. Cancer Imaging (2023) 23:67 Page 2 of 8 Background morphological characteristics in small LNs (short-axis Patients with locally advanced rectal cancer (LARC) diameter < 5 mm) have been considered challenging [10]. and those who are unable to undergo radical resection In 2016, a new criterion was added to European Society are recommend to receive neoadjuvant therapy, which of Gastrointestinal and Abdominal Radiology (ESGAR), aims to increase the possibility of complete resection considering LNs with a short-axis diameter < 5  mm as and reduce the risk of local recurrence by downstaging benign after neoadjuvant therapy [6]. To the best of and downsizing [1]. After neoadjuvant therapy, most of our knowledge, no study has validated this criterion for the lymph nodes (LNs) shrink and disappear, but the LNs a node-by-node evaluation. Thus, this study aimed to that can still be detected on MRI carry a risk of metasta- validate the ESGAR criterion and explore the diagnos- sis, with LNs larger than 5  mm having a 38.6% possibil- tic efficiency of nodal morphological characteristics for ity of being pathologically metastatic [2]. Even in patients LNs ≥ 5 mm. with a pathological complete response, 3.2-14% of vis- ible LNs are malignant [3, 4]. Malignant LNs can lead to Methods poor prognosis, including local recurrence and reduced Patients long-term survival rate [1, 5]. Therefore, for patients after This retrospective study was approved by our Institu - neoadjuvant therapy, particularly whose who are nearly tional Review Board, and written informed consent was clinical complete responders, MRI restaging of LNs obtained from all patients. A total of 262 consecutive is valuable in evaluating the feasibility of conservative patients with LARC treated with neoadjuvant therapy treatment, such as a local excision or a watch-and-wait between March 2015 and June 2021 were considered strategy. for inclusion. Among these patients, 172 patients were Restaging of benign and malignant LNs is more chal- excluded for the following reasons (a) no total mesorectal lenging than primary staging in rectal cancer. Most stud- excision 6–8 weeks after neoadjuvant therapy (n = 108); ies recognize that MRI criteria for LNs involvement in (b) no post-treatment MRI within 2 weeks before surgery primary staging include size, shape, border, signal homo- (n = 30); (c) poor image quality for evaluation (n = 12); (d) geneity, and enhancement homogeneity [6–8]. However, no LNs found on MRI (n = 20); and (e) the location of the these criteria are not entirely suitable for nodal restating LNs on MRI did not match the histopathologic results after neoadjuvant therapy [9, 10]. Most research has pri- (n = 2). Finally, 90 patients (68 men, 22 women; median marily relied on the size criterion and has not considered age: 55 years, range: 20–90 years) were included in this whether morphological characteristics (including shape, study (Fig. 1). border, signal homogeneity and enhancement homo- geneity) can effectively distinguish benign and malig - Neoadjuvant therapy nant LNs [4, 9, 11]. Only a few studies have shown that Neoadjuvant therapy included preoperative chemora- nodal morphological characteristics contribute to nodal diotherapy and preoperative chemotherapy. Preoperative restaging [12–14]. However, evaluating the details of chemoradiotherapy regimen involved long-term radio- therapy using volumetric modulated arc therapy with Fig. 1 Flow diagram of patients included Zhang et al. Cancer Imaging (2023) 23:67 Page 3 of 8 a total dose of 50–60  Gy for primary and nodal gross homogeneity (homogeneous/ heterogeneous) on CE- tumor volumes administered with conventional seg- 3D-HR-T1WI in consensus. mentation, along with concurrent chemotherapy. Che- motherapy regimens included the mFOLFOX regimen Radiologic–histopathologic comparison (leucovorin + fluorouracil + oxaliplatin), CapeOX regimen All visible regional LNs on HR-T2WI were divided into (capecitabine + oxaliplatin), or capecitabine monotherapy. three groups, including mesorectal, superior rectal and inferior mesenteric LNs. Based on the agreed-upon MRI examination nodal position by the radiologist and surgeon, an expe- Patients, except those with lower and large rectal tumors, rienced surgeon specializing in colorectal cancer succes- were infused with an appropriate amount (20–80 mL) of sively localized and removed the regional LNs in different ultrasonic gel into the rectum. To reduce bowel motility, groups during surgery. In order to provide an accurate 20  mg of raceanisodamine hydrochloride was injected node-by-node comparison between MR images and his- intramuscularly 10  min before MRI examination unless topathological findings, special attention was paid to contraindicated. nodal size and morphology, as well as nodal relative posi- MRI was performed using a 3.0-T scanner (Magnetom tion to the tumor, rectal wall, mesorectal fascia, vessels Verio, Siemens Healthcare) with a 6-channel phased- and adjacent LNs [15]. The excised LNs were then sent to array body coil. Patients were positioned supine with feet the pathology department and quickly placed in individ- first. The rectal MRI protocols included (a) sagittal, coro - ual trays. All LNs were fixed in formalin and stained with nal and oblique axial (orthogonal to tumor base) non-fat- hematoxylin and eosin. Thereafter, an experienced gas - suppressed high-spatial-resolution T2- weighted imaging trointestinal pathologist analyzed and classified each LN (HR-T2WI) using a turbo spin-echo sequence; and (b) as benign or malignant under light microscope. The LNs coronal fat-suppressed isotropic contrast-enhanced reported by histopathology were rematched with HR- three-dimensional high-spatial-resolution T1-weighted T2WI in the corresponding groups and were excluded if imaging (CE-3D-HR-T1WI) using a gradient-echo they could not be matched. sequence. Axial, sagittal and coronal multiplanar recon- structions were performed on CE-3D-HR-T1WI with a Statistical analysis slice thickness of 3 mm. An intravenous bolus of 0.2 mL/ Statistical analyses were performed using SPSS software kg gadopentetate dimeglumine was injected at a rate of (version 25.0, IBM). Figures were generated using Graph- 3.0 mL/s, followed by a 25-mL saline flush at the same Pad Prism (version 9.0, GraphPad Software) and Med- rate. Detailed protocols are listed in Table 1. Calc statistical software (version 15.8, MedCalc Software bvba, Ostend, Belgium). The normality of quantitative Image interpretation data was test by using Kolmogorov-Smirnov test. Nor- Two radiologists with 2 and 7 years of experience in rec- mally distributed data were compared with the indepen- tal MRI, reviewed all MR images blinded to histopatho- dent samples t-test, while nonnormally distributed data logic findings. Firstly, all visible LNs on HR-T2WI were were presented as medians with ranges and compared determined by the two radiologists in consensus. Sec- with the Mann-Whitney U test. Categorical data were ondly, the short-axis diameters of LNs were measured expressed as numbers with percentages and compared independently, and the average values were calculated with the χ test or Fisher’s exact test. ROC curve was con- for subsequent analyses. Thirdly, the two radiologists structed, and the AUC with a 95% confidence interval assessed the shape (oval/round), border (smooth/irregu- was calculated to evaluate the diagnostic efficacy of nodal lar), and signal homogeneity (homogeneous/heteroge- size. With a short-axis diameter cutoff value of 5 mm, the neous) of LNs ≥ 5  mm on HR-T2WI and enhancement sensitivity, specificity, positive predictive value (PPV), Table 1 High-spatial-resolution MRI protocols for rectal cancer Sequences TR/TE (ms) Slice thickness/ Slices Base resolution Phase FOV (mm) Voxel size (mm ) Acquisi- Gap (mm) resolution tion time (%) High-spatial-resolution T2-weighted imaging Sagittal 3000/87 3/0 19 320 80 180 0.7 × 0.6 × 3.0 2 min 30 s Coronal 4000/77 3/0 25 384 80 220 0.7 × 0.6 × 3.0 2 min 52 s Oblique axial 3000/84 3/0 24 320 100 180 0.6 × 0.6 × 3.0 3 min 18 s Contrast-enhanced three-dimensional high-spatial-resolution T1-weighted imaging Coronal 10/4.9 1/0.2 144 384 100 380 1.0 × 1.0 × 1.0 3 min 10 s TR, repetition time; TE, echo time; FOV, field of view Zhang et al. Cancer Imaging (2023) 23:67 Page 4 of 8 negative predictive value (NPV) and accuracy were cal- malignant LNs were 65.9% (29/44), 87.0% (221/254), culated. The morphological characteristics were com - 46.8% (29/62), 93.6% (221/236) and 83.9% (250/298), pared for LNs ≥ 5  mm using the χ test or Fisher’s exact respectively (Table  3). Among LNs < 5  mm, almost all test. A two-tailed P value < 0.05 indicated a statistically (93.6%, 221/236) were benign, while less than half (46.8%, significant difference. 29/62) LNs ≥ 5 mm were malignant (Fig. 3). Results Nodal status compared with its morphological Clinicopathologic findings characteristics A total of 90 patients were included, of which 27 (30.0%, Among the 62 LNs ≥ 5  mm, 21.2% (7/33) of benign LNs 27/90) were confirmed to have malignant LNs. Malig - showed an oval shape, while 86.2% (25/29) of malignant nant LNs were more likely occur in patients with pT3-4 LNs were round. Smooth borders were observed in 69.7% tumors (P < 0.05) (Table  2). Histopathology of 1049 (23/33) of benign LNs, compared to only 17.2% (5/29) LNs harvested from the rectal specimens in 90 patients of malignant LNs with irregular borders. Homogeneous (median:11, range: 0–31) revealed that 72 (6.9%, 72/1049) signal intensity was present in 66.7% (22/33) of benign were malignant. A total of 308 LNs were visualized on LNs, while 41.4% (12/29) of malignant LNs showed het- MRI in the 90 patients (median:3, range:1–8). For a node- erogeneous signal intensity. Furthermore, 21.2% (7/33) of by-node evaluation, 298 LNs were successfully matched benign LNs exhibited homogeneous enhancement, com- between MRI and histopathology, of which 44 were pared to 86.2% (25/29) of malignant LNs with hetero- malignant. geneous enhancement (Fig.  4). However, none of these morphological characteristics were found to be statisti- Diagnostic value of nodal size cally significant in differentiating benign and malignant The median short-axis diameter was 3.7  mm (range: LNs (P > 0.05) (Table 4). 1.6-9  mm) for the 254 benign LNs and 5.9  mm (range:2.4–13  mm) for the 44 malignant LNs. The AUC Discussion was 0.81 (95% confidence interval: 0.74–0.89), indicat - For the evaluation of LNs in rectal cancer, many studies ing that nodal size had moderate diagnostic performance have focused on assessing the agreement between MRI in distinguishing malignant from benign LNs (Fig.  2). and histopathology for N-staging [16, 17]. Although this When LNs ≥ 5  mm were considered malignant, sensitiv- method is easy to implement, it ignores the characteris- ity, specificity, PPV, NPV and accuracy for determining tics of each individual node. In our study, we conducted Table 2 Relationship between clinicopathologic features and lymph nodes metastases in 90 patients Parameters Total ypN0 ypN+ P value (n = 90) (n = 63) (n = 27) * * * a Age (years) 55 (24–82) 56(27–82) 55(24–75) 0.926 Sex 0.830 Man 68 (75.6%) 48(76.2%) 20(74.1%) Woman 22 (24.4%) 15(23.8%) 7(25.9%) # c Tumor location 0.777 Upper 11(12.2%) 7(11.1%) 4(14.8%) Middle 51(56.7%) 35(55.6%) 16(59.3%) Lower 28(31.1%) 21(33.3%) 7(25.9%) Histological subtype 0.234 Nonmucinous adenocarcinoma 82(91.1%) 59(93.7%) 23(85.2%) Mucinous adenocarcinoma 8(8.9%) 4(6.3%) 4(14.8.%) Differentiation 0.808 Well 1(1.1%) 1(1.6%) 0(0.0%) Moderate 78(86.7%) 55(87.3%) 23(85.2%) Poor 11(12.2%) 7(11.1%) 4(14.8%) ypT stage 0.004 ypT0-2 44(48.9%) 37(58.7%) 7(25.9%) ypT3-4 46(51.1%) 26(41.3%) 20(74.1%) Date are medians and ranges in parentheses According to the distance from the most caudal border of the rectal tumor to the anal verge on MRI: upper, > 10 cm; middle, 5–10 cm; lower, < 5 cm a b 2 c Mann-Whitney U test, χ test, Fisher’s exact test yp, post-neoadjuvant treatment pathological feature Zhang et al. Cancer Imaging (2023) 23:67 Page 5 of 8 Fig. 2 ROC curve showing the diagnostic performance of nodal size for malignant lymph nodes a node-by-node analysis, matching and verifying each Many studies have shown that the number of LNs usually LN found on MRI with the pathological results. This decreases, and the short-axis diameter becomes smaller approach has been proven to be more specific and valu - or even disappears after neoadjuvant therapy. Downsized able in establishing diagnostic criteria for LNs [7, 18]. LNs have a very low chance of metastasis [2]. However, Our study validated that the ESGAR criterion had a there is still no consensus on the optimal cutoff value. In moderate diagnostic performance (AUC = 0.81) for nodal the 2016 ESGAR, a new item was added for MRI nodal restaging in rectal cancer after neoadjuvant therapy. With restaging: all LNs < 5  mm should be considered benign, a short-axis diameter cutoff value of 5  mm, the sensi - but there are no reliable criteria for LNs ≥ 5  mm. Our tivity, specificity and accuracy were 65.9%, 87.0% and study also showed that short-axis diameter cutoff value 83.9%, respectively. The size criterion is the most widely of 5  mm had a high NPV, as almost all (93.6%, 221/236) used indicator for nodal staging and restaging in cur- LNs < 5  mm were benign. Thus, we can confidently con - rent studies on rectal cancer LNs [9, 19]. According to clude that there are no malignant LNs if all LNs on MRI the ESGAR consensus meeting in 2012, the size criterion after neoadjuvant therapy are < 5  mm. However, among after neoadjuvant therapy is more reliable than the base- the 62 LNs ≥ 5  mm, malignant LNs accounted for only line MRI assessment for diagnosing malignant LNs [19]. 46.8% (29/62), resulting a low PPV. Therefore, it is impos - sible to effectively distinguish benign and malignant LNs ≥ 5 mm. Table 3 Comparison of nodal size and histopathologic findings Although we used HR-T2WI and CE-3D-HR-T1WI in 298 lymph nodes with thin thickness in our study, confidently observing Histopathologic the morphological features of small LNs remains chal- findings lenging [9, 10]. Consequently, we only evaluated the Nodal size ypLN- ypLN+ Total LN- (< 5 mm) 221 15 236 morphological features of LNs ≥ 5  mm. Although mor- LN+ (≥ 5 mm) 33 29 62 phological characteristics were effective in predicting Total 254 44 298 the status of LNs before treatment [18], we found no sig- LNs, lymph nodes; yp, post-neoadjuvant treatment pathological feature nificant differences in shape, border, signal homogeneity, Zhang et al. Cancer Imaging (2023) 23:67 Page 6 of 8 Fig. 3 Short-axis diameter distribution of benign and malignant LNs. Dotted line: mean, solid line: cutoff value. 93.6% of the LNs < 5 mm were benign (green spots), and only 46.8% of the LNs ≥ 5 mm were malignant (red spots). LNs, lymph nodes and enhancement homogeneity between benign and Conclusions malignant LNs after neoadjuvant therapy. Nodal shape In summary, our study demonstrates that the ESGAR has is likely influenced by the scanning planes. Even benign moderate diagnostic performance for nodal restaging LNs showed irregular border and heterogeneous signal in rectal cancer after neoadjuvant therapy. LNs < 5  mm intensity. This could be attributed to fibrous thickening of can be effectively identified as benign using the size cri - the capsule and fibrotic or mucinous changes after treat - terion alone, in line with the ESGAR recommendation. ment, which make elevation more complex [20]. Regard- However, morphological features do not aid in the diag- ing enhancement homogeneity evaluation, we applied nosis of LNs ≥ 5  mm on MRI restaging. Future research fat-suppressed isotropic CE-3D-HR-T1WI, which has should focus on refining the criteria for distinguishing demonstrated good performance in nodal staging [7]. benign and malignant LNs ≥ 5 mm and exploring alterna- Compared with HR-T2WI, this sequence offers advan - tive imaging techniques to improve the accuracy of nodal tages such as higher spatial resolution, improved signal- restaging in rectal cancer after neoadjuvant therapy. to-noise ratio, and multiplanar reconstruction, providing better-detailed of LNs. However, heterogenous enhance- ment could still be observed in benign LNs ≥ 5 mm, likely due to nodal fibrosis or the presence of acellular mucin lakes caused by neoadjuvant therapy [20, 21]. Malignant LNs with smooth border, homogeneous signal intensity, and enhancement may contain micrometastases. There were some limitations in our study. Firstly, we only evaluated the morphological features of LNs ≥ 5 mm because accurately assessing the morphological features of small LNs is challenging. Secondly, the number of metastatic LNs was lower than that of benign because patients with obvious LN metastasis usually do not undergo surgery. Finally, we did not evaluate iliac LNs as extended pelvic lymphadenectomy is not usually per- formed in total mesorectal excision. Zhang et al. Cancer Imaging (2023) 23:67 Page 7 of 8 Fig. 4 Coronal non-fat-suppressed high-spatial-resolution T2-weighted imaging (a, c and e); coronal fat-suppressed isotropic contrast-enhanced three- dimensional high-spatial-resolution T1-weighted imaging (b, d and f ). The lymph nodes were signed by white boxes. a-b, Benign node with 3.7 mm in short-axis diameter, showed oval, smooth border, homogeneous signal and homogeneous enhancement. c-d, Benign node with 5.6 mm in short-axis diameter, showed round, irregular border, heterogeneous signal and heterogeneous enhancement. e-f, Malignant node with 5.8 mm in short-axis diam- eter, showed round, irregular border, heterogeneous signal and heterogeneous enhancement List of abbreviations Table 4 Comparison of nodal morphological characteristics and LARC locally advanced rectal cancer histopathologic findings in 62 lymph nodes ≥ 5 mm LNs lymph nodes ESGAR European Society of Gastrointestinal and Abdominal Histopathologic Radiology findings HR-T2WI high-spatial-resolution T2-weighted imaging Morphological characteristics ypLN- ypLN+ P value CE-3D-HR-T1WI contrast-enhanced three-dimensional high-spatial- (n = 33) (n = 29) resolution T1-weighted imaging Shape 0.445 PPV positive predictive value NPV negative predictive value Oval 7 (21.2%) 4 (13.8%) Round 26 (78.8%) 25 (86.2%) Acknowledgements Border 0.231 None. Smooth 23 (69.7%) 24 (82.8%) Authors’ contributions Irregular 10 (30.3%) 5 (17.2%) ZZ and YC contributed equally to this work. SY put forward the study concepts Signal homogeneity 0.513 and ZZ designed the study. Data acquisition and interpretation were done by Homogeneous 22 (66.7%) 17 (58.6%) all authors. ZZ and YC wrote and edited the manuscript. All authors read and approved the final manuscript. Heterogeneous 11 (33.3%) 12 (41.4%) Enhancement homogeneity 0.445 Funding Homogeneous 7 (21.2%) 4 (13.8%) Support by Guangdong Basic and Applied Basic Research Foundation (2020A1515010796). Heterogeneous 26 (78.8%) 25 (86.2%) LNs, lymph nodes; yp, post-neoadjuvant treatment pathological feature Data Availability a 2 χ test The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Zhang et al. Cancer Imaging (2023) 23:67 Page 8 of 8 chemoradiation–can we rely on radiologic nodal staging after chemoradia- Declarations tion? Dis Colon Rectum. 2009;52(7):1278–84. 10. Seo N, Kim H, Cho MS, Lim JS. Response Assessment with MRI after Ethics approval and consent to participate Chemoradiotherapy in rectal Cancer: current evidences. Korean J Radiol. This study was approved by our institutional review board. Written informed 2019;20(7):1003–18. consent was obtained from each patient. 11. Kim DJ, Kim JH, Lim JS, Yu JS, Chung JJ, Kim MJ, et al. Restaging of rectal Can- cer with MR Imaging after Concurrent Chemotherapy and Radiation Therapy. Consent for publication Radiographics. 2010;30(2):503–16. Not applicable. 12. Barbaro B, Fiorucci C, Tebala C, Valentini V, Gambacorta MA, Vecchio FM, et al. Locally advanced rectal cancer: MR imaging in prediction of response Competing interests after preoperative chemotherapy and radiation therapy. Radiology. The authors declare that they have no competing interests. 2009;250(3):730–9. 13. Nahas SC, Nahas CSR, Cama GM, de Azambuja RL, Horvat N, Marques CFS, Received: 12 February 2023 / Accepted: 2 July 2023 et al. Diagnostic performance of magnetic resonance to assess treatment response after neoadjuvant therapy in patients with locally advanced rectal cancer. Abdom Radiol (NY ). 2019;44(11):3632–40. 14. Almlov K, Woisetschlager M, Loftas P, Hallbook O, Elander NO, Sandstrom P. MRI lymph node evaluation for prediction of metastases in rectal Cancer. Anticancer Res. 2020;40(5):2757–63. References 15. Cho EY, Kim SH, Yoon JH, Lee Y, Lim YJ, Kim SJ, et al. Apparent diffusion coef - 1. Glynne-Jones R, Wyrwicz L, Tiret E, Brown G, Rodel C, Cervantes A, et al. Rectal ficient for discriminating metastatic from non-metastatic lymph nodes in cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow- primary rectal cancer. Eur J Radiol. 2013;82(11):E662–E8. up. Ann Oncol. 2017;28(suppl4):iv22–iv40. 16. Grone J, Loch FN, Taupitz M, Schmidt C, Kreis ME. Accuracy of various Lymph 2. Heijnen LA, Maas M, Beets-Tan RG, Berkhof M, Lambregts DM, Nelemans PJ, et Node staging criteria in rectal Cancer with magnetic resonance imaging. J al. Nodal staging in rectal cancer: why is restaging after chemoradiation more Gastrointest Surg. 2018;22(1):146–53. accurate than primary nodal staging? Int J Colorectal Dis. 2016;31(6):1157–62. 17. Koh DM, Chau I, Tait D, Wotherspoon A, Cunningham D, Brown G. Evaluating 3. Ryan JE, Warrier SK, Lynch AC, Heriot AG. Assessing pathological complete mesorectal lymph nodes in rectal cancer before and after neoadjuvant response to neoadjuvant chemoradiotherapy in locally advanced rectal chemoradiation using thin-section T2-weighted magnetic resonance imag- cancer: a systematic review. Colorectal Dis. 2015;17(10):849–61. ing. Int J Radiat Oncol Biol Phys. 2008;71(2):456–61. 4. Lahaye MJ, Beets GL, Engelen SM, Kessels AG, de Bruine AP, Kwee HW, et al. 18. Brown G, Richards CJ, Bourne MW, Newcombe RG, Radcliffe AG, Dallimore Locally advanced rectal cancer: MR imaging for restaging after neoadjuvant NS, et al. Morphologic predictors of lymph node status in rectal cancer with radiation therapy with concomitant chemotherapy. Part II. What are the use of high-spatial-resolution MR imaging with histopathologic comparison. criteria to predict involved lymph nodes? Radiology. 2009;252(1):81–91. Radiology. 2003;227(2):371–7. 5. Dinaux AM, Leijssen L, Bordeianou LG, Kunitake H, Amri R, Berger DL. Out- 19. Beets-Tan RG, Lambregts DM, Maas M, Bipat S, Barbaro B, Caseiro-Alves F, et comes of persistent lymph node involvement after neoadjuvant therapy for al. Magnetic resonance imaging for the clinical management of rectal cancer stage III rectal cancer. Surgery. 2018;163(4):784–8. patients: recommendations from the 2012 european Society of Gastroin- 6. Beets-Tan RGH, Lambregts DMJ, Maas M, Bipat S, Barbaro B, Curvo-Semedo L, testinal and Abdominal Radiology (ESGAR) consensus meeting. Eur Radiol. et al. Magnetic resonance imaging for clinical management of rectal cancer: 2013;23(9):2522–31. updated recommendations from the 2016 european Society of Gastroin- 20. Prall F, Wohlke M, Klautke G, Schiffmann L, Fietkau R, Barten M. Tumour testinal and Abdominal Radiology (ESGAR) consensus meeting. Eur Radiol. regression and mesorectal lymph node changes after intensified 2018;28(4):1465–75. neoadjuvant chemoradiation for carcinoma of the rectum. APMIS. 7. Chen Y, Yang X, Wen Z, Lu B, Xiao X, Shen B, et al. Fat-suppressed gadolinium- 2006;114(3):201–10. enhanced isotropic high-resolution 3D-GRE-T1WI for predicting small node 21. Chetty R, McCarthy AJ. Neoadjuvant chemoradiation and rectal cancer. J Clin metastases in patients with rectal cancer. Cancer Imaging. 2018;18(1):21. Pathol. 2019;72(2):97–101. 8. Jao SY, Yang BY, Weng HH, Yeh CH, Lee LW. Evaluation of gadolinium- enhanced T1-weighted magnetic resonance imaging in the preop- erative assessment of local staging in rectal cancer. Colorectal Dis. Publisher’s Note 2010;12(11):1139–48. Springer Nature remains neutral with regard to jurisdictional claims in 9. Perez RO, Pereira DD, Proscurshim I, Gama-Rodrigues J, Rawet V, Sao published maps and institutional affiliations. Juliao GP, et al. Lymph node size in rectal cancer following neoadjuvant http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cancer Imaging Springer Journals

MRI for nodal restaging after neoadjuvant therapy in rectal cancer with histopathologic comparison

Download PDF
8 pages

Loading next page...
 
/lp/springer-journals/mri-for-nodal-restaging-after-neoadjuvant-therapy-in-rectal-cancer-KcsA2hBz6F

References (43)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2023
eISSN
1470-7330
DOI
10.1186/s40644-023-00589-0
Publisher site
See Article on Publisher Site

Abstract

Background After neoadjuvant therapy, most of the lymph nodes (LNs) will shrink and disappear in patients with rectal cancer. However, LNs that are still detectable on MRI carry a risk of metastasis. This study aimed to evaluate the performance of the European Society of Gastrointestinal and Abdominal Radiology (ESGAR) criterion (short-axis diameter ≥ 5 mm) in diagnosing malignant LNs in patients with rectal cancer after neoadjuvant therapy, and whether nodal morphological characteristics (including shape, border, signal homogeneity, and enhancement homogeneity) could improve the diagnostic efficiency for LNs ≥ 5 mm. Methods This retrospective study included 90 patients with locally advanced rectal cancer who underwent surgery after neoadjuvant therapy and performed preoperative MRI. Two radiologists independently measured the short- axis diameter of LNs and evaluated the morphological characteristics of LNs ≥ 5 mm in consensus. With a per node comparison with histopathology as the reference standard, a ROC curve was performed to evaluate the diagnostic performance of the size criterion. For categorical variables, either a χ test or Fisher’s exact test was used. Results A total of 298 LNs were evaluated. The AUC for nodal size in determining nodal status was 0.81. With a size cutoff value of 5 mm, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 65.9%, 87.0%, 46.8%, 93.6% and 83.9%, respectively. No significant differences were observed in any of the morphological characteristics between benign and malignant LNs ≥ 5 mm (all P > 0.05). Conclusions The ESGAR criterion demonstrated moderate diagnostic performance in identifying malignant LNs in patients with rectal cancer after neoadjuvant therapy. It was effective in determining the status of LNs < 5 mm but not for LNs ≥ 5 mm, and the diagnostic efficiency could not be improved by considering nodal morphological characteristics. Keywords Rectal cancer, Neoadjuvant therapy, Lymph node, Magnetic resonance imaging, Histopathology *Correspondence: Shenping Yu yushp@mail.sysu.edu.cn Department of Radiology, The First Affiliated Hospital, Sun Yat-sen University, 510080 Guangzhou, China Department of Radiology, The Seventh Affiliated Hospital, Sun Yat-sen University, 518036 Shenzhen, China MR Scientific Marketing, SIEMENS Healthineers Ltd, 200124 Shanghai, China © The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhang et al. Cancer Imaging (2023) 23:67 Page 2 of 8 Background morphological characteristics in small LNs (short-axis Patients with locally advanced rectal cancer (LARC) diameter < 5 mm) have been considered challenging [10]. and those who are unable to undergo radical resection In 2016, a new criterion was added to European Society are recommend to receive neoadjuvant therapy, which of Gastrointestinal and Abdominal Radiology (ESGAR), aims to increase the possibility of complete resection considering LNs with a short-axis diameter < 5  mm as and reduce the risk of local recurrence by downstaging benign after neoadjuvant therapy [6]. To the best of and downsizing [1]. After neoadjuvant therapy, most of our knowledge, no study has validated this criterion for the lymph nodes (LNs) shrink and disappear, but the LNs a node-by-node evaluation. Thus, this study aimed to that can still be detected on MRI carry a risk of metasta- validate the ESGAR criterion and explore the diagnos- sis, with LNs larger than 5  mm having a 38.6% possibil- tic efficiency of nodal morphological characteristics for ity of being pathologically metastatic [2]. Even in patients LNs ≥ 5 mm. with a pathological complete response, 3.2-14% of vis- ible LNs are malignant [3, 4]. Malignant LNs can lead to Methods poor prognosis, including local recurrence and reduced Patients long-term survival rate [1, 5]. Therefore, for patients after This retrospective study was approved by our Institu - neoadjuvant therapy, particularly whose who are nearly tional Review Board, and written informed consent was clinical complete responders, MRI restaging of LNs obtained from all patients. A total of 262 consecutive is valuable in evaluating the feasibility of conservative patients with LARC treated with neoadjuvant therapy treatment, such as a local excision or a watch-and-wait between March 2015 and June 2021 were considered strategy. for inclusion. Among these patients, 172 patients were Restaging of benign and malignant LNs is more chal- excluded for the following reasons (a) no total mesorectal lenging than primary staging in rectal cancer. Most stud- excision 6–8 weeks after neoadjuvant therapy (n = 108); ies recognize that MRI criteria for LNs involvement in (b) no post-treatment MRI within 2 weeks before surgery primary staging include size, shape, border, signal homo- (n = 30); (c) poor image quality for evaluation (n = 12); (d) geneity, and enhancement homogeneity [6–8]. However, no LNs found on MRI (n = 20); and (e) the location of the these criteria are not entirely suitable for nodal restating LNs on MRI did not match the histopathologic results after neoadjuvant therapy [9, 10]. Most research has pri- (n = 2). Finally, 90 patients (68 men, 22 women; median marily relied on the size criterion and has not considered age: 55 years, range: 20–90 years) were included in this whether morphological characteristics (including shape, study (Fig. 1). border, signal homogeneity and enhancement homo- geneity) can effectively distinguish benign and malig - Neoadjuvant therapy nant LNs [4, 9, 11]. Only a few studies have shown that Neoadjuvant therapy included preoperative chemora- nodal morphological characteristics contribute to nodal diotherapy and preoperative chemotherapy. Preoperative restaging [12–14]. However, evaluating the details of chemoradiotherapy regimen involved long-term radio- therapy using volumetric modulated arc therapy with Fig. 1 Flow diagram of patients included Zhang et al. Cancer Imaging (2023) 23:67 Page 3 of 8 a total dose of 50–60  Gy for primary and nodal gross homogeneity (homogeneous/ heterogeneous) on CE- tumor volumes administered with conventional seg- 3D-HR-T1WI in consensus. mentation, along with concurrent chemotherapy. Che- motherapy regimens included the mFOLFOX regimen Radiologic–histopathologic comparison (leucovorin + fluorouracil + oxaliplatin), CapeOX regimen All visible regional LNs on HR-T2WI were divided into (capecitabine + oxaliplatin), or capecitabine monotherapy. three groups, including mesorectal, superior rectal and inferior mesenteric LNs. Based on the agreed-upon MRI examination nodal position by the radiologist and surgeon, an expe- Patients, except those with lower and large rectal tumors, rienced surgeon specializing in colorectal cancer succes- were infused with an appropriate amount (20–80 mL) of sively localized and removed the regional LNs in different ultrasonic gel into the rectum. To reduce bowel motility, groups during surgery. In order to provide an accurate 20  mg of raceanisodamine hydrochloride was injected node-by-node comparison between MR images and his- intramuscularly 10  min before MRI examination unless topathological findings, special attention was paid to contraindicated. nodal size and morphology, as well as nodal relative posi- MRI was performed using a 3.0-T scanner (Magnetom tion to the tumor, rectal wall, mesorectal fascia, vessels Verio, Siemens Healthcare) with a 6-channel phased- and adjacent LNs [15]. The excised LNs were then sent to array body coil. Patients were positioned supine with feet the pathology department and quickly placed in individ- first. The rectal MRI protocols included (a) sagittal, coro - ual trays. All LNs were fixed in formalin and stained with nal and oblique axial (orthogonal to tumor base) non-fat- hematoxylin and eosin. Thereafter, an experienced gas - suppressed high-spatial-resolution T2- weighted imaging trointestinal pathologist analyzed and classified each LN (HR-T2WI) using a turbo spin-echo sequence; and (b) as benign or malignant under light microscope. The LNs coronal fat-suppressed isotropic contrast-enhanced reported by histopathology were rematched with HR- three-dimensional high-spatial-resolution T1-weighted T2WI in the corresponding groups and were excluded if imaging (CE-3D-HR-T1WI) using a gradient-echo they could not be matched. sequence. Axial, sagittal and coronal multiplanar recon- structions were performed on CE-3D-HR-T1WI with a Statistical analysis slice thickness of 3 mm. An intravenous bolus of 0.2 mL/ Statistical analyses were performed using SPSS software kg gadopentetate dimeglumine was injected at a rate of (version 25.0, IBM). Figures were generated using Graph- 3.0 mL/s, followed by a 25-mL saline flush at the same Pad Prism (version 9.0, GraphPad Software) and Med- rate. Detailed protocols are listed in Table 1. Calc statistical software (version 15.8, MedCalc Software bvba, Ostend, Belgium). The normality of quantitative Image interpretation data was test by using Kolmogorov-Smirnov test. Nor- Two radiologists with 2 and 7 years of experience in rec- mally distributed data were compared with the indepen- tal MRI, reviewed all MR images blinded to histopatho- dent samples t-test, while nonnormally distributed data logic findings. Firstly, all visible LNs on HR-T2WI were were presented as medians with ranges and compared determined by the two radiologists in consensus. Sec- with the Mann-Whitney U test. Categorical data were ondly, the short-axis diameters of LNs were measured expressed as numbers with percentages and compared independently, and the average values were calculated with the χ test or Fisher’s exact test. ROC curve was con- for subsequent analyses. Thirdly, the two radiologists structed, and the AUC with a 95% confidence interval assessed the shape (oval/round), border (smooth/irregu- was calculated to evaluate the diagnostic efficacy of nodal lar), and signal homogeneity (homogeneous/heteroge- size. With a short-axis diameter cutoff value of 5 mm, the neous) of LNs ≥ 5  mm on HR-T2WI and enhancement sensitivity, specificity, positive predictive value (PPV), Table 1 High-spatial-resolution MRI protocols for rectal cancer Sequences TR/TE (ms) Slice thickness/ Slices Base resolution Phase FOV (mm) Voxel size (mm ) Acquisi- Gap (mm) resolution tion time (%) High-spatial-resolution T2-weighted imaging Sagittal 3000/87 3/0 19 320 80 180 0.7 × 0.6 × 3.0 2 min 30 s Coronal 4000/77 3/0 25 384 80 220 0.7 × 0.6 × 3.0 2 min 52 s Oblique axial 3000/84 3/0 24 320 100 180 0.6 × 0.6 × 3.0 3 min 18 s Contrast-enhanced three-dimensional high-spatial-resolution T1-weighted imaging Coronal 10/4.9 1/0.2 144 384 100 380 1.0 × 1.0 × 1.0 3 min 10 s TR, repetition time; TE, echo time; FOV, field of view Zhang et al. Cancer Imaging (2023) 23:67 Page 4 of 8 negative predictive value (NPV) and accuracy were cal- malignant LNs were 65.9% (29/44), 87.0% (221/254), culated. The morphological characteristics were com - 46.8% (29/62), 93.6% (221/236) and 83.9% (250/298), pared for LNs ≥ 5  mm using the χ test or Fisher’s exact respectively (Table  3). Among LNs < 5  mm, almost all test. A two-tailed P value < 0.05 indicated a statistically (93.6%, 221/236) were benign, while less than half (46.8%, significant difference. 29/62) LNs ≥ 5 mm were malignant (Fig. 3). Results Nodal status compared with its morphological Clinicopathologic findings characteristics A total of 90 patients were included, of which 27 (30.0%, Among the 62 LNs ≥ 5  mm, 21.2% (7/33) of benign LNs 27/90) were confirmed to have malignant LNs. Malig - showed an oval shape, while 86.2% (25/29) of malignant nant LNs were more likely occur in patients with pT3-4 LNs were round. Smooth borders were observed in 69.7% tumors (P < 0.05) (Table  2). Histopathology of 1049 (23/33) of benign LNs, compared to only 17.2% (5/29) LNs harvested from the rectal specimens in 90 patients of malignant LNs with irregular borders. Homogeneous (median:11, range: 0–31) revealed that 72 (6.9%, 72/1049) signal intensity was present in 66.7% (22/33) of benign were malignant. A total of 308 LNs were visualized on LNs, while 41.4% (12/29) of malignant LNs showed het- MRI in the 90 patients (median:3, range:1–8). For a node- erogeneous signal intensity. Furthermore, 21.2% (7/33) of by-node evaluation, 298 LNs were successfully matched benign LNs exhibited homogeneous enhancement, com- between MRI and histopathology, of which 44 were pared to 86.2% (25/29) of malignant LNs with hetero- malignant. geneous enhancement (Fig.  4). However, none of these morphological characteristics were found to be statisti- Diagnostic value of nodal size cally significant in differentiating benign and malignant The median short-axis diameter was 3.7  mm (range: LNs (P > 0.05) (Table 4). 1.6-9  mm) for the 254 benign LNs and 5.9  mm (range:2.4–13  mm) for the 44 malignant LNs. The AUC Discussion was 0.81 (95% confidence interval: 0.74–0.89), indicat - For the evaluation of LNs in rectal cancer, many studies ing that nodal size had moderate diagnostic performance have focused on assessing the agreement between MRI in distinguishing malignant from benign LNs (Fig.  2). and histopathology for N-staging [16, 17]. Although this When LNs ≥ 5  mm were considered malignant, sensitiv- method is easy to implement, it ignores the characteris- ity, specificity, PPV, NPV and accuracy for determining tics of each individual node. In our study, we conducted Table 2 Relationship between clinicopathologic features and lymph nodes metastases in 90 patients Parameters Total ypN0 ypN+ P value (n = 90) (n = 63) (n = 27) * * * a Age (years) 55 (24–82) 56(27–82) 55(24–75) 0.926 Sex 0.830 Man 68 (75.6%) 48(76.2%) 20(74.1%) Woman 22 (24.4%) 15(23.8%) 7(25.9%) # c Tumor location 0.777 Upper 11(12.2%) 7(11.1%) 4(14.8%) Middle 51(56.7%) 35(55.6%) 16(59.3%) Lower 28(31.1%) 21(33.3%) 7(25.9%) Histological subtype 0.234 Nonmucinous adenocarcinoma 82(91.1%) 59(93.7%) 23(85.2%) Mucinous adenocarcinoma 8(8.9%) 4(6.3%) 4(14.8.%) Differentiation 0.808 Well 1(1.1%) 1(1.6%) 0(0.0%) Moderate 78(86.7%) 55(87.3%) 23(85.2%) Poor 11(12.2%) 7(11.1%) 4(14.8%) ypT stage 0.004 ypT0-2 44(48.9%) 37(58.7%) 7(25.9%) ypT3-4 46(51.1%) 26(41.3%) 20(74.1%) Date are medians and ranges in parentheses According to the distance from the most caudal border of the rectal tumor to the anal verge on MRI: upper, > 10 cm; middle, 5–10 cm; lower, < 5 cm a b 2 c Mann-Whitney U test, χ test, Fisher’s exact test yp, post-neoadjuvant treatment pathological feature Zhang et al. Cancer Imaging (2023) 23:67 Page 5 of 8 Fig. 2 ROC curve showing the diagnostic performance of nodal size for malignant lymph nodes a node-by-node analysis, matching and verifying each Many studies have shown that the number of LNs usually LN found on MRI with the pathological results. This decreases, and the short-axis diameter becomes smaller approach has been proven to be more specific and valu - or even disappears after neoadjuvant therapy. Downsized able in establishing diagnostic criteria for LNs [7, 18]. LNs have a very low chance of metastasis [2]. However, Our study validated that the ESGAR criterion had a there is still no consensus on the optimal cutoff value. In moderate diagnostic performance (AUC = 0.81) for nodal the 2016 ESGAR, a new item was added for MRI nodal restaging in rectal cancer after neoadjuvant therapy. With restaging: all LNs < 5  mm should be considered benign, a short-axis diameter cutoff value of 5  mm, the sensi - but there are no reliable criteria for LNs ≥ 5  mm. Our tivity, specificity and accuracy were 65.9%, 87.0% and study also showed that short-axis diameter cutoff value 83.9%, respectively. The size criterion is the most widely of 5  mm had a high NPV, as almost all (93.6%, 221/236) used indicator for nodal staging and restaging in cur- LNs < 5  mm were benign. Thus, we can confidently con - rent studies on rectal cancer LNs [9, 19]. According to clude that there are no malignant LNs if all LNs on MRI the ESGAR consensus meeting in 2012, the size criterion after neoadjuvant therapy are < 5  mm. However, among after neoadjuvant therapy is more reliable than the base- the 62 LNs ≥ 5  mm, malignant LNs accounted for only line MRI assessment for diagnosing malignant LNs [19]. 46.8% (29/62), resulting a low PPV. Therefore, it is impos - sible to effectively distinguish benign and malignant LNs ≥ 5 mm. Table 3 Comparison of nodal size and histopathologic findings Although we used HR-T2WI and CE-3D-HR-T1WI in 298 lymph nodes with thin thickness in our study, confidently observing Histopathologic the morphological features of small LNs remains chal- findings lenging [9, 10]. Consequently, we only evaluated the Nodal size ypLN- ypLN+ Total LN- (< 5 mm) 221 15 236 morphological features of LNs ≥ 5  mm. Although mor- LN+ (≥ 5 mm) 33 29 62 phological characteristics were effective in predicting Total 254 44 298 the status of LNs before treatment [18], we found no sig- LNs, lymph nodes; yp, post-neoadjuvant treatment pathological feature nificant differences in shape, border, signal homogeneity, Zhang et al. Cancer Imaging (2023) 23:67 Page 6 of 8 Fig. 3 Short-axis diameter distribution of benign and malignant LNs. Dotted line: mean, solid line: cutoff value. 93.6% of the LNs < 5 mm were benign (green spots), and only 46.8% of the LNs ≥ 5 mm were malignant (red spots). LNs, lymph nodes and enhancement homogeneity between benign and Conclusions malignant LNs after neoadjuvant therapy. Nodal shape In summary, our study demonstrates that the ESGAR has is likely influenced by the scanning planes. Even benign moderate diagnostic performance for nodal restaging LNs showed irregular border and heterogeneous signal in rectal cancer after neoadjuvant therapy. LNs < 5  mm intensity. This could be attributed to fibrous thickening of can be effectively identified as benign using the size cri - the capsule and fibrotic or mucinous changes after treat - terion alone, in line with the ESGAR recommendation. ment, which make elevation more complex [20]. Regard- However, morphological features do not aid in the diag- ing enhancement homogeneity evaluation, we applied nosis of LNs ≥ 5  mm on MRI restaging. Future research fat-suppressed isotropic CE-3D-HR-T1WI, which has should focus on refining the criteria for distinguishing demonstrated good performance in nodal staging [7]. benign and malignant LNs ≥ 5 mm and exploring alterna- Compared with HR-T2WI, this sequence offers advan - tive imaging techniques to improve the accuracy of nodal tages such as higher spatial resolution, improved signal- restaging in rectal cancer after neoadjuvant therapy. to-noise ratio, and multiplanar reconstruction, providing better-detailed of LNs. However, heterogenous enhance- ment could still be observed in benign LNs ≥ 5 mm, likely due to nodal fibrosis or the presence of acellular mucin lakes caused by neoadjuvant therapy [20, 21]. Malignant LNs with smooth border, homogeneous signal intensity, and enhancement may contain micrometastases. There were some limitations in our study. Firstly, we only evaluated the morphological features of LNs ≥ 5 mm because accurately assessing the morphological features of small LNs is challenging. Secondly, the number of metastatic LNs was lower than that of benign because patients with obvious LN metastasis usually do not undergo surgery. Finally, we did not evaluate iliac LNs as extended pelvic lymphadenectomy is not usually per- formed in total mesorectal excision. Zhang et al. Cancer Imaging (2023) 23:67 Page 7 of 8 Fig. 4 Coronal non-fat-suppressed high-spatial-resolution T2-weighted imaging (a, c and e); coronal fat-suppressed isotropic contrast-enhanced three- dimensional high-spatial-resolution T1-weighted imaging (b, d and f ). The lymph nodes were signed by white boxes. a-b, Benign node with 3.7 mm in short-axis diameter, showed oval, smooth border, homogeneous signal and homogeneous enhancement. c-d, Benign node with 5.6 mm in short-axis diameter, showed round, irregular border, heterogeneous signal and heterogeneous enhancement. e-f, Malignant node with 5.8 mm in short-axis diam- eter, showed round, irregular border, heterogeneous signal and heterogeneous enhancement List of abbreviations Table 4 Comparison of nodal morphological characteristics and LARC locally advanced rectal cancer histopathologic findings in 62 lymph nodes ≥ 5 mm LNs lymph nodes ESGAR European Society of Gastrointestinal and Abdominal Histopathologic Radiology findings HR-T2WI high-spatial-resolution T2-weighted imaging Morphological characteristics ypLN- ypLN+ P value CE-3D-HR-T1WI contrast-enhanced three-dimensional high-spatial- (n = 33) (n = 29) resolution T1-weighted imaging Shape 0.445 PPV positive predictive value NPV negative predictive value Oval 7 (21.2%) 4 (13.8%) Round 26 (78.8%) 25 (86.2%) Acknowledgements Border 0.231 None. Smooth 23 (69.7%) 24 (82.8%) Authors’ contributions Irregular 10 (30.3%) 5 (17.2%) ZZ and YC contributed equally to this work. SY put forward the study concepts Signal homogeneity 0.513 and ZZ designed the study. Data acquisition and interpretation were done by Homogeneous 22 (66.7%) 17 (58.6%) all authors. ZZ and YC wrote and edited the manuscript. All authors read and approved the final manuscript. Heterogeneous 11 (33.3%) 12 (41.4%) Enhancement homogeneity 0.445 Funding Homogeneous 7 (21.2%) 4 (13.8%) Support by Guangdong Basic and Applied Basic Research Foundation (2020A1515010796). Heterogeneous 26 (78.8%) 25 (86.2%) LNs, lymph nodes; yp, post-neoadjuvant treatment pathological feature Data Availability a 2 χ test The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Zhang et al. Cancer Imaging (2023) 23:67 Page 8 of 8 chemoradiation–can we rely on radiologic nodal staging after chemoradia- Declarations tion? Dis Colon Rectum. 2009;52(7):1278–84. 10. Seo N, Kim H, Cho MS, Lim JS. Response Assessment with MRI after Ethics approval and consent to participate Chemoradiotherapy in rectal Cancer: current evidences. Korean J Radiol. This study was approved by our institutional review board. Written informed 2019;20(7):1003–18. consent was obtained from each patient. 11. Kim DJ, Kim JH, Lim JS, Yu JS, Chung JJ, Kim MJ, et al. Restaging of rectal Can- cer with MR Imaging after Concurrent Chemotherapy and Radiation Therapy. Consent for publication Radiographics. 2010;30(2):503–16. Not applicable. 12. Barbaro B, Fiorucci C, Tebala C, Valentini V, Gambacorta MA, Vecchio FM, et al. Locally advanced rectal cancer: MR imaging in prediction of response Competing interests after preoperative chemotherapy and radiation therapy. Radiology. The authors declare that they have no competing interests. 2009;250(3):730–9. 13. Nahas SC, Nahas CSR, Cama GM, de Azambuja RL, Horvat N, Marques CFS, Received: 12 February 2023 / Accepted: 2 July 2023 et al. Diagnostic performance of magnetic resonance to assess treatment response after neoadjuvant therapy in patients with locally advanced rectal cancer. Abdom Radiol (NY ). 2019;44(11):3632–40. 14. Almlov K, Woisetschlager M, Loftas P, Hallbook O, Elander NO, Sandstrom P. MRI lymph node evaluation for prediction of metastases in rectal Cancer. Anticancer Res. 2020;40(5):2757–63. References 15. Cho EY, Kim SH, Yoon JH, Lee Y, Lim YJ, Kim SJ, et al. Apparent diffusion coef - 1. Glynne-Jones R, Wyrwicz L, Tiret E, Brown G, Rodel C, Cervantes A, et al. Rectal ficient for discriminating metastatic from non-metastatic lymph nodes in cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow- primary rectal cancer. Eur J Radiol. 2013;82(11):E662–E8. up. Ann Oncol. 2017;28(suppl4):iv22–iv40. 16. Grone J, Loch FN, Taupitz M, Schmidt C, Kreis ME. Accuracy of various Lymph 2. Heijnen LA, Maas M, Beets-Tan RG, Berkhof M, Lambregts DM, Nelemans PJ, et Node staging criteria in rectal Cancer with magnetic resonance imaging. J al. Nodal staging in rectal cancer: why is restaging after chemoradiation more Gastrointest Surg. 2018;22(1):146–53. accurate than primary nodal staging? Int J Colorectal Dis. 2016;31(6):1157–62. 17. Koh DM, Chau I, Tait D, Wotherspoon A, Cunningham D, Brown G. Evaluating 3. Ryan JE, Warrier SK, Lynch AC, Heriot AG. Assessing pathological complete mesorectal lymph nodes in rectal cancer before and after neoadjuvant response to neoadjuvant chemoradiotherapy in locally advanced rectal chemoradiation using thin-section T2-weighted magnetic resonance imag- cancer: a systematic review. Colorectal Dis. 2015;17(10):849–61. ing. Int J Radiat Oncol Biol Phys. 2008;71(2):456–61. 4. Lahaye MJ, Beets GL, Engelen SM, Kessels AG, de Bruine AP, Kwee HW, et al. 18. Brown G, Richards CJ, Bourne MW, Newcombe RG, Radcliffe AG, Dallimore Locally advanced rectal cancer: MR imaging for restaging after neoadjuvant NS, et al. Morphologic predictors of lymph node status in rectal cancer with radiation therapy with concomitant chemotherapy. Part II. What are the use of high-spatial-resolution MR imaging with histopathologic comparison. criteria to predict involved lymph nodes? Radiology. 2009;252(1):81–91. Radiology. 2003;227(2):371–7. 5. Dinaux AM, Leijssen L, Bordeianou LG, Kunitake H, Amri R, Berger DL. Out- 19. Beets-Tan RG, Lambregts DM, Maas M, Bipat S, Barbaro B, Caseiro-Alves F, et comes of persistent lymph node involvement after neoadjuvant therapy for al. Magnetic resonance imaging for the clinical management of rectal cancer stage III rectal cancer. Surgery. 2018;163(4):784–8. patients: recommendations from the 2012 european Society of Gastroin- 6. Beets-Tan RGH, Lambregts DMJ, Maas M, Bipat S, Barbaro B, Curvo-Semedo L, testinal and Abdominal Radiology (ESGAR) consensus meeting. Eur Radiol. et al. Magnetic resonance imaging for clinical management of rectal cancer: 2013;23(9):2522–31. updated recommendations from the 2016 european Society of Gastroin- 20. Prall F, Wohlke M, Klautke G, Schiffmann L, Fietkau R, Barten M. Tumour testinal and Abdominal Radiology (ESGAR) consensus meeting. Eur Radiol. regression and mesorectal lymph node changes after intensified 2018;28(4):1465–75. neoadjuvant chemoradiation for carcinoma of the rectum. APMIS. 7. Chen Y, Yang X, Wen Z, Lu B, Xiao X, Shen B, et al. Fat-suppressed gadolinium- 2006;114(3):201–10. enhanced isotropic high-resolution 3D-GRE-T1WI for predicting small node 21. Chetty R, McCarthy AJ. Neoadjuvant chemoradiation and rectal cancer. J Clin metastases in patients with rectal cancer. Cancer Imaging. 2018;18(1):21. Pathol. 2019;72(2):97–101. 8. Jao SY, Yang BY, Weng HH, Yeh CH, Lee LW. Evaluation of gadolinium- enhanced T1-weighted magnetic resonance imaging in the preop- erative assessment of local staging in rectal cancer. Colorectal Dis. Publisher’s Note 2010;12(11):1139–48. Springer Nature remains neutral with regard to jurisdictional claims in 9. Perez RO, Pereira DD, Proscurshim I, Gama-Rodrigues J, Rawet V, Sao published maps and institutional affiliations. Juliao GP, et al. Lymph node size in rectal cancer following neoadjuvant

Journal

Cancer Imaging – Springer Journals

Published: Jul 13, 2023

Keywords: Rectal cancer; Neoadjuvant therapy; Lymph node; Magnetic resonance imaging; Histopathology

There are no references for this article.