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

Learn More →

The national incidence of chondrosarcoma of bone; a review

The national incidence of chondrosarcoma of bone; a review ACTA ONCOLOGICA https://doi.org/10.1080/0284186X.2023.2177975 REVIEW a,b b,c Joachim Thorkildsen and Tor Åge Myklebust a b Division of Orthopaedic Surgery, Oslo University Hospital, Nydalen, Oslo, Norway; Department of Registration, Cancer Registry of Norway, Majorstuen, Oslo, Norway; Department of Research, Møre & Romsdal Hospital Trust, Ålesund, Norway ABSTRACT ARTICLE HISTORY Received 11 January 2023 Background: Chondrosarcoma (CS) epidemiology has been studied by a number of authors using Accepted 1 February 2023 national cancer registry cohorts. Many reports share the common findings of a slight increase in inci- dence, but not all. The patterns and causes for these changes are divergent while reflection concern- KEYWORDS ing methodological challenges are often missing. Chondrosarcoma; incidence; Method: We have performed a structured literature review to find national analyses of CS incidence review; epidemiology published from 2010 to 2020. We included eight studies of national incidence of CS, summarise their findings and patterns of change. We further discuss explanations given for these changes to better understand the real patterns and raise awareness in their interpretation. Results: Reported crude incidence ranges from 0.27 per million per year overall in Saudi Arabia to 5.4 in the Netherlands. Four studies from the USA, England, Switzerland and France report age standar- dised rates of 2.0–4.1 per million per year overall. While some countries report stable patterns, most report a slight increase. The Netherlands is the only country reporting a large increase, driven by a 10- fold increase in the incidence of ACT/grade 1 CS during the study period. We challenge the explana- tions given for this and suggest that this most likely is a result of variable interpretation and definition of CS at the lower levels of disease aggressiveness. This should raise awareness to possible over-treat- ment of CS in the Netherlands. Conclusion: The most likely national incidence of CS of bone is between 2–4 per million per year. Three modern reports present an incidence of 3.4–4.1 per million per year. Background Method The Norwegian Cancer Registry’s (NCR) main objective is to We have followed the PRISMA guidelines for review as far as establish knowledge and disseminate information that con- possible. Inclusion criteria are national cohorts with crude or age-standardised incidence rates (ASR), published in 2010– tributes to reducing cancer (https://www.kreftregisteret.no/ 2020. Exclusion criteria are institutional data sets, data for en). With this mandate, we published a national analysis of limited sites, subtypes or patient groups and prognostic incidence and prognostication for Chondrosarcoma (CS) of reports without incidence data. bone in Norway in 2019 [1]. The literature review and selection of articles are illus- A number of new articles with similar methodology, from trated in Figure 1. The main author completed a ‘Pubmed’ other countries, have been published [2,3] since then and a search for the terms ‘chondrosarcoma and incidence’ on the new edition of the WHO book has been released [4]. The 21.05.21. The search was limited to publication dates 2010– WHO state that the incidence of low-grade CS has increased 2020. This gave 445 results. The main author screened the over the past decade as a result of the improved sensitivity title and abstract or full text where necessary. of radiological scans and refers to a national epidemiological A total of 20 papers were identified as being epidemio- study from the Netherlands [5]. They do so without compara- logical studies including an assessment of the incidence of tive analysis, while the incidence reported in the Netherlands chondrosarcoma and were reviewed in full text. Assessment is by far the highest in the world. Furthermore, an inter- of these 20 papers led to the inclusion of eight national epi- national comparison of bone sarcoma epidemiology found demiological studies from the USA [7], England [7], The that CS was the bone sarcoma type with the most variability Netherlands [5], Norway [8], Switzerland [3], France [2], in reporting [6]. Taiwan[9] and Saudi Arabia[10]. A systematic analysis and comparison are therefore The other 12 were excluded. One study contained data for required to establish what the correct incidence of CS of certain grades of disease only[11] while another for specific bone is. Further, to explore potential causes of change and age range of patients[12]. Five studies were limited to regional variability in reporting. [10,13–17] and 4 institutional [18–21] data. One review paper CONTACT Joachim Thorkildsen jthork@ous-hf.no Division of Orthopaedic Surgery, Oslo University Hospital, PO Box 4956, Nydalen, 0424, Oslo, Norway This study was performed at Oslo University Hospital, Oslo. The interpretation and reporting of these data are the sole responsibility of the authors. 2023 Acta Oncologica Foundation 2 J. THORKILDSEN AND T. Å. MYKLEBUST Figure 1. Flow chart for literature review and inclusion of studies. was excluded though it contained data from multiple nations per year, they are converted to rates per million per year for through the International Agency for Research on Cancer [6]. ease of comparison. The overall crude incidence rate varied from 0.27 per million This is a database for records on cancer incidence based on in Saudi Arabia to 5.4 per million in the Netherlands. Four both regional and national registries, but it does not list the European reports in modern time periods find the crude inci- methodology at the individual registries. dence of 2.85, 4.2, 5.0 and 5.4 overall. Three studies report age We extracted data for crude and age-standardised inci- standardised rates with the European standard population from dence (ASR) rates by sex and overall, together with the 2.0–4.1 per million/year while three reports using the world standard population used for calculation from text, tables standard publish ASR of 1.2–3.4 per million per year. The high- and supplementary material published. We present only the est incidence of any type published in subgroups or time peri- extracted data in their original form. Table 1, summarises the ods was 8.78 in 2005–2013 for the Netherlands. Five nations features of the included studies. (USA, England, Netherlands, Norway and France) report an increase in the incidence of CS over the study period, whilst the Swiss study found no evidence of change. In Taiwan and Results Saudi Arabia, the patterns of change were not examined. The results from the included studies are summarised in Five studies demonstrated higher incidence (crude/ASR) Table 2. Where ASR’s are presented per 100,000 inhabitants values for men than for women, while in the USA the rate ACTA ONCOLOGICA 3 Table 1. Summary of included studies ASR ¼ Age standardised rate. Surname main Incidence Author / year Inclusion Nation of type Population of publication period origin Register name Register type reported standard used Arnfinsen 2011 1976–2005 USA Surveillance, Epidemiology, and Combination of 9 regional ASR World End Results program (SEER) cancer registries Whelan 2012 1998–2007 England National Cancer Data Combination of 8 regional ASR European Repository (NCDR) cancer registries Hung 2014 2003–2010 Taiwan Taiwan Cancer Registry (TCR) National cancer registry ASR World van Praag 2018 1989–2013 Netherlands Netherlands Cancer Registry National cancer registry Crude None (NCR-IKNL) Thorkildsen 2019 1990–2013 Norway Cancer Registry of National cancer registry Crude None Norway (CRN) Kollar 2019 1996–2015 Switzerland National Institute for Cancer Combination of Swiss Crude ASR European Epidemiology and Cantonal Cancer Registration (NICER) Registries Amadeo 2020 2000–2013 France French Network of cancer Combination of 19 regional Crude ASR European, Registries (FRANCIM) cancer registries world, US Aljuhani 2020 2013-2017 Saudi Arabia Saudi Cancer Registry (SCR) National cancer registry Crude None Table 2. Summary of results from included studies. Surname main Crude Highest Time Explanation Author rate ASR Incidence (type) trend proposed Arnfinsen 2011 Na 2.7 men 2.7 (ASRw) Increase Exogenous estrogen 2.7 women Whelan 2012 Na 2.6 men 2.0 (ASRe) Increase, then Change in reporting 1.7 women stabilisation 2.0 total Hung 2014 Na 1.34 men 1.2 (ASRw) Not examined Na 1.02 women 1.2 total van Praag 2018 5,4 overall Na 8.78 (Crude) Increase Increase in ACT due to increased imaging and aging population Thorkildsen 2019 3.04 men Na 3.45 (Crude) Increase None given 2.67 women 2.85 overall Kollar 2019 5,0 men 4.3 men 3.4 (ASRe) No change Na 3.5 women 3.0 women 4.2 total 3.4 total Amadeo 2020 5.0 men 4.1 men 4.1 (ASRe) Increase Exogenous estrogen 4.0 women 3.6 women 5.0 total 4.1 total Aljuhani 2020 0.27 total Na 0.27 (Crude) Not examined Na a b Arnfinsen reports over period 1976–2005, but in three time periods. The most recent time period features her is 1996–2005. ASR given for European, world and US reference. European is used here. ASRw: Age standardised rate by world standard population, ASRe: Age standardised rate by European standard popu- lation. Na:not available. was similar between the sexes. Three studies included data The Dutch present a crude overall incidence of 2.88 per on incidence over time, by sex. In England, Whelan et al. million per year for 1989–96 increasing to 8.78 for 2005–13. showed an increased ASR for both sexes (37% for men and The increase is driven by a rise in the incidence of 40% for women), mostly occurring during the 1980s with sta- ACT/grade 1 CS (2013 WHO definition), with an increase bilization thereafter. In the USA however, there was no from 1.2 per million in 1989–96 to 6.63 in 2005–13. change in the ASR incidence among men, but an increase Extracting data from their Figure 2 shows a crude incidence for women through the study period of 1976–2005. In for ACT/grade 1 CS in 2011 and 2012 of 9.8 per million per Norway, there was an increase in crude incidence for both year. Van Praag et al. also show a doubling of the incidence sexes (23% for men and 82% for women). The increase for of grade 2 disease from 0.68 per million per year1989–96 to women was such that the rate was equal, compared to men 1.38 per million per year in 2005–13. Grade 3 disease in the final segment of the study (2009–2013). increases from 0.27 per million per year in 1989–96 to 0.41 The studies from Norway and the Netherlands both have per million per year in 1997–04, but then remains stable at more detailed data at the subtype level allowing us to look 0.43 per million per year in 2005–2013. These last two peri- beyond sex. The increase in Norway is driven by a rise in the ods show an incidence of grade 3 CS remarkably similar to incidence of the central CS subtype, which increases with the Norwegian numbers. age. The increase was evident for grade 1 and 2 diseases For the most modern time period, excluding the while grade 3 disease remained stable. The increase was Netherlands, the incidence of CS was 3.4 per million per year largest for grade 2 disease. for Switzerland (ASR-europe), 3.45 for Norway (Crude) and 4 J. THORKILDSEN AND T. Å. MYKLEBUST Figure 2. Incidence of ACT/CS grade 1 in the Netherlands, between 1889 and 2013, number of MRI examinations over time in the Netherlands. Adapted from van Praag et al. with addition of number of MRI examinations in Norway. 4.1 for France (ASR-Europe). Similar findings, from similar Five authors present rates by ASR, but another methodical periods of time and countries, support the validity of the dimension is that ASR can be calculated with a range of dif- findings. ferent standard populations as presented in Table 1. Different age groups are weighted differently in different ref- erence populations. As such, this can give significant differ- Discussion ences. Two authors use the world standard, two use the European standard and one uses multiple standards. This has In a global assessment of bone sarcoma incidence from 68 been discussed in the annual Norwegian Cancer Registry countries reporting to the International Agency for Reporting report of 2019 in relation to their usage of the Norwegian of Cancer (IARC) for 2003–2007, only small changes in inci- Standard [22]. As illustrated by Figure 3, the Norwegian dence over time are noted for CS [6]. Although increases standard weights the elderly population significantly more have been reported for both men and women in the than the world standard. The overall national cancer ASR per Netherlands [5] and Norway [1], the IARC report shows 100,000 per year for all sites in Norway is 700 for men by equivalent decreases in CS rates among men in Sweden and the Norwegian standard while only 360 by the world stand- women in Ireland and Japan.[6] ard. For ‘Bone’ location however it is 1.2 for the Norwegian Valery et al. also describe the largest variability of report- standard while 1.0 by the world standard. Amadeo et al. ing for CS, compared to Ewing and Osteosarcoma. CS repre- illustrate this by presenting their numbers with both. The sents <10% of reported bone sarcomas in India and Saudi French national incidence is 20% higher if calculated by the Arabia, while >45% in Finland, Slovenia and the Netherlands. European standard (4.1 per million per year) as compared to They also state that although CS incidence rose with age in the world standard (3.4 per million per year) [2]. We have all countries, a peak was seen earlier and more pronounced attempted to list these different standards and methods in countries with higher incidence. Varied patterns of report- clearly in both text and tables to allow for comparison and ing will naturally lead to variability in estimated incidence. raise awareness. Although an important point of accuracy, There are some important methodical challenges to an this difference can not explain the size of the differences accurate comparison of incidence. First, key publications seen between the Dutch numbers at the upper end and the from Norway and the Netherlands include crude incidence Saudi Arabian at the lower end. only. These are figures for the number of cases of CS over a With regards to methods of reporting, the articles period of time divided by the population at risk in the nation included also represent different types of registries. Norway, in the same study time without adjustment. Both these pub- the Netherlands and Taiwan have a single national registry. lications of national cohorts present crude incidence despite both national registries from which they are drawn routinely The UK, France and Switzerland have a national registry that using ASR to present incidence. ASR is observed age-specific combines data from regional registries with presumed rates applied to a standard population and adjusted by a national complete coverage. weight depending on the composition of the standard popu- The USA has a registry that also combines data from lation used. They thereby give a rate adjusted for age, allow- regional registries, but which covers 30% of the US popula- ing comparison between groups and over time. tion. They then extrapolate data based on this. A recent ACTA ONCOLOGICA 5 Figure 3. Comparison of population weights for Norwegian and world standard reference populations from Cancer Registry of Norway. assessment of evidence-based CS staging from the US noted this basis, it is likely that differences in access to-, and that up to 40% of skeletal CS entries in the SEER database resources of the health care system correlate to the amount lacked essential variable information [23]. In SEER publica- of CS discovered. In particular, it is likely as van Praag et al. tions, up to 10% of cases included are termed myxoid chon- argue, that the frequency of use of diagnostic radiology serv- drosarcoma [24,25]This is not a separate subtype in the WHO ices would likely correlate to the volume of CS discovered. categorization of chondrogenic sarcoma, but rather myxoid The worldbank publishes statistics for current health change is a part of pathological grading. It is possible that expenditure per capita in US$(https://data.worldbank.org/). these cases in fact represent extraskeletal myxoid sarcoma The values of US$per capita spending in 2019 range from which is a soft tissue sarcoma with a misleading name since $10 623 for the USA at the upper end, to $1316 for Saudi these are not organised under chondrogenic differentiation, Arabia at the lower end. Taiwan is not listed in the statistics. but rather uncertain differentiation by the WHO. There is a These numbers confirm that there is a substantial difference publication of an exceedingly rare CS subtype termed myx- in the health care systems of the countries included and this oid CS or Chordoid sarcoma [26], but it seems unlikely that will likely represent a certain difference in the use of diag- this represents 10% of a national cohort not reported by nostic radiology and thereby CS diagnosis, but also in meth- other countries. ods and resources used in reporting cancer. Norway has mandatory reporting of CS to the register by A further unique feature of CS is its well-established law for all personnel involved in cancer care since 1952. inter-and intra-observer variability in assessment by radiol- The Netherlands also has mandatory reporting by law. The ogists, pathologists and clinicians [31–33]. This is most evi- Cancer Control Act of 2003 in Taiwan also obligates all hos- dent at the lower level of disease aggressiveness, in pitals with greater than 50-bed capacity, involved in cancer distinguishing between benign enchondroma and grade 1 care, to participate in reporting all newly diagnosed malig- CS or ACT. At the same time, central chondroid lesions of nant neoplasms [27]They estimate that 80% of cancer the skeleton appear in approximately 2–3% of routinely per- patients are diagnosed and treated in such units. The formed radiological examinations of the knee and shoulder Norwegian register has published rates of completeness [34,35]. There is no uniform accepted practice for the assess- [28], but similarly does Taiwan [27]. A comparative analysis ment or follow-up of these lesions. It is methodically likely of the national cancer registries of the Nordic countries of that this spectrum of disease is central to any difference in reporting. Norway, Sweden, Finland, Denmark and Iceland concluded The lower numbers in Taiwan and Saudi Arabia can of that although these were more similar than any other col- lection of data from five different countries there were sig- course also be an expression of ethnic difference in inci- nificant differences in registration, classification and dence as suggested by a comparative analysis of Japanese, inclusion [29]. Chinese, and US bone sarcoma cases [36]. An epidemio- A key feature of CS epidemiology is that the overall prog- logical study from New York in the USA found that the nosis for CS is better and that tumour growth is more indo- reported incidence for CS was lower for the ethnic black lent than other primary bone sarcomas. It has been shown population as compared to the ethnic white population [37]. that 6% of diagnosed CS were asymptomatic [30] while in They argued that this partially can be explained by the black the more recent Norwegian study, as much as 15% of CS population having shorter life expectancy and therefore was retrospectively categorised as ‘incidental findings’ [1]. On lower at-risk exposure in the elderly age groups where CS is 6 J. THORKILDSEN AND T. Å. MYKLEBUST more common. Ethnic difference is however not a finding The Dutch increase was driven by a massive rise in found for CS by an international comparative analysis of ACT/CS grade 1, not experienced by comparable countries. bone cancer epidemiology [6]. Histological grading has not formally changed during the Two articles quote exogenous estrogens as a likely associ- period. At the same time this is the entity most open to ation with increased levels of CS. The study of CS and estro- interobserver variation. The natural explanation for their gens is however divergent. Although estrogen signaling has increase is one of variable definitions and a more careful been shown to be active in cartilaginous tumours [38], the practice. It seems likely that the Netherlands have a lower study of inhibiting estrogen in vitro and in vivo showed no level of interpretation and intervention for central chondroid beneficial effect [39]. In addition, Laitinen et al. showed that lesions compared to other countries. In the assessment of a being hormonally active appears to give survival advantages central chondroid lesion they are more likely to conclude in women with CS [40]. with the likelihood of ACT/grade 1 CS than enchondroma, The two most similar reports come from Norway and the compared to other nations. The scale of this appears to be Netherlands. Both studies provide subtype-specific crude quite large. We cannot presume to know the cause or reason incidence data over a similar modern time period from well- for this practice. Whatever the reason, it most probably rep- functioning and well-resourced national registries in similar resents a significant level of over-treatment. This is sup- societies and populations. All the same, they report vastly ported by the publication of safe observation without different results, driven by different subgroups of disease. treatment of extremity grade 1 CS/ACT [43] in the Van Praag et al. offer an explanation of the increased use of Netherlands. There is no cause in the literature to suspect diagnostic images and an aging population. Both of these under-treatment of these entities. Even with minimally inva- explanations can be further examined. sive procedures, CS management is associated with consider- In Norway, the Norwegian Radiation and Nuclear Safety able morbidity [44–46]. Authority provide interim reports on the use of radiological examinations at a national and international level[41]. They demonstrate in their reports from 2002 and 2008 that the Conclusion rates of radiological investigations performed was slightly Removing the outliers it appears that the incidence of CS is higher in Norway compared to the Netherlands in one report likely to be in the range of 2-4 cases per million inhabitants and conversely in the other. The number of MRI performed per year. Two reports using similar population standards pre- per 1000 inhabitants in Norway per year was 61 in 2002 and sent ASR rates of 3.4 and 4.1 per million per year. A majority 126 in 2008 [43]. These numbers can then be entered into van Praag’s adapted figure (Figure 2) as red stars. Provided of countries report a slight increase. This may in part be that a proportion of CS are incidental findings, an increase in explained by increased use of diagnostic imaging. The level use of MRI can support a limited increase in CS incidence, of CS reported by the Netherlands driven by increasing but not of the same magnitude as that reported from the ACT/grade 1 CS is uniquely high. The most likely explanation Netherlands. Furthermore, with similar use of MRI in Norway is one of variable definitions of this lesion and possible over- and the Netherlands, the change in CS incidence should also treatment. be similar, which it is not. This lends supports to the notion that the same lesions are probably interpreted, managed Disclosure statement and reported differently in the two nations. It is known that the incidence of CS rises with age [1,7,42] No potential conflict of interest was reported by the author(s). and as such the notion that an aging population can explain an increase in CS incidence is logical. Population data for the Funding Netherlands is readily available from the Statistics Netherlands (http://www.cbs.nl/en-gb/figures/detail/37556eng). There was None to disclose an increase of 33% in the proportion of the population older than 65 years of age from 1990–2010, or 632,000 people in ORCID total. These numbers are far too small in absolute terms to explain a 10-fold rise in incidence of grade 1 CS which is a Tor Åge Myklebust http://orcid.org/0000-0003-4645-1635 rare tumor even when using the reported Dutch estimates. This can, only to a limited degree, explain a rise in the inci- Data availability statement dence of CS. Further, if they used age standardised rates, rather than crude rates, this change in population would All dara are available from the main author upon reasonable request. already be adjusted for in their rate. The Netherlands is a leading voice in CS research and a References defining force for CS terminology through the WHO. The findings and explanations of van Praag et al. are reproduced [1] Thorkildsen J, Taksdal I, Bjerkehagen B, et al. Chondrosarcoma th in the 5 edition WHO book [4], but in our opinion with the in Norway 1990–2013; an epidemiological and prognostic insufficient investigation. The estimated numbers from the observational study of a complete national cohort. Acta Oncol. Netherlands need further explanation. 2019;58(3):1–10. ACTA ONCOLOGICA 7 [2] Amadeo B, Penel N, Coindre JM, et al. Incidence and time trends from the SEER database. J Bone Joint Surg Am. 2009;91(5): of sarcoma (2000–2013): results from the French network of can- 1063–1072. cer registries (FRANCIM). BMC Cancer. 2020;20(1):190. [25] Amer KM, Munn M, Congiusta D, et al. Survival and prognosis of [3] Kollar A, Rothermundt C, Klenke F, et al. Incidence, mortality, and chondrosarcoma subtypes: SEER database analysis. J Orthop Res. survival trends of soft tissue and bone sarcoma in Switzerland 2020;38(2):311–319. between 1996 and 2015. Cancer Epidemiol. 2019;63:101596. [26] Kilpatrick SE, Inwards CY, Fletcher CD, et al. Myxoid chondrosar- [4] WHO Classification of Tumours Editorial Board. Soft tissue and coma (chordoid sarcoma) of bone: a report of two cases and bone tumours. 5th ed. Board WCE, editor. Lyon (France): review of the literature. Cancer. 1997;79(10):1903–1910. International Agency for Research on Cancer; 2020. [27] Chiang CJ, You SL, Chen CJ, et al. Quality assessment and [5] van Praag VM, Rueten-Budde AJ, Ho V, et al. Incidence, outcomes improvement of nationwide cancer registration system in Taiwan: and prognostic factors during 25 years of treatment of chondro- a review. Jpn J Clin Oncol. 2015;45(3):291–296. sarcomas. Surg Oncol. 2018;27(3):402–408. [28] Larsen IK, Smastuen M, Johannesen TB, et al. Data quality at the [6] Valery PC, Laversanne M, Bray F. Bone cancer incidence by mor- cancer registry of Norway: an overview of comparability, com- phological subtype: a global assessment. Cancer Causes Control. pleteness, validity and timeliness. Eur J Cancer. 2009;45(7):1218– 2015;26(8):1127–1139. [7] Whelan J, McTiernan A, Cooper N, et al. Incidence and survival of [29] Pukkala E, Engholm G, Højsgaard Schmidt LK, et al. Nordic cancer malignant bone sarcomas in England 1979–2007. Int J Cancer. registries - an overview of their procedures and data comparabil- 2012;131(4):E508–E517. ity. Acta Oncol. 2018;57(4):440–455. [8] Thorkildsen J, Taksdal I, Bjerkehagen B, et al. Chondrosarcoma in [30] BjoR € nsson J, McLeod RA, Unni KK, et al. Primary chondrosar- Norway 1990–2013; an epidemiological and prognostic observa- coma of long bones and limb girdles. Cancer. 1998;83(10):2105– tional study of a complete national cohort. Acta Oncol. 2019; 58(3):273–282. [31] Eefting D, Schrage YM, Geirnaerdt MJ, et al. Assessment of inter- [9] Hung GY, Horng JL, Yen HJ, et al. Incidence patterns of primary observer variability and histologic parameters to improve reliabil- bone cancer in Taiwan (2003–2010): a population-based study. ity in classification and grading of Central cartilaginous tumors. Ann Surg Oncol. 2014;21(8):2490–2498. Am J Surg Pathol. 2009;33(1):50–57. [10] Aljuhani WS, Alanazi AM, Alghafees MA. Primary bone sarcomas [32] Skeletal Lesions Interobserver Correlation among Expert in KSA: a Saudi tumor registry review. J Taibah Univ Medical Sci. Diagnosticians (SLICED) Study Group Reliability of histopathologic 2021;16(1):77–85. and radiologic grading of cartilaginous neoplasms in long bones. [11] Goedhart LM, Ho VKY, Dijkstra PDS, et al. Bone sarcoma incidence J Bone Joint Surg Am. 2007;89:2113–2123. in The Netherlands. Cancer Epidemiol. 2019;60:31–38. [33] Zamora T, Urrutia J, Schweitzer D, et al. Do orthopaedic oncolo- [12] Eyre R, Feltbower RG, James PW, et al. The epidemiology of bone gists agree on the diagnosis and treatment of cartilage tumors of cancer in 0–39 year olds in Northern England, 1981 - 2002. BMC the appendicular skeleton? Clin Orthop Relat Res. 2017;475(9): Cancer. 2010;10(1):357. 2176–2186. [13] Ducimetiere F, Lurkin A, Ranchere-Vince D[, et al. Incidence rate, [34] Stomp W, Reijnierse M, Kloppenburg M, et al. Prevalence of cartil- epidemiology of sarcoma and molecular biology. Preliminary aginous tumours as an incidental finding on MRI of the knee. Eur results from EMS study in the Rhone-Alpes region. Bull Cancer. Radiol. 2015;25(12):3480–3487. 2010;97:629–641. [35] Hong ED, Carrino JA, Weber KL, et al. Prevalence of shoulder [14] Solooki S, Vosoughi AR, Masoomi V. Epidemiology of musculo- enchondromas on routine MR imaging. Clin Imaging. 2011;35(5): skeletal tumors in shiraz, South of Iran. Indian J Med Paediatr 378–384. Oncol. 2011;32(4):187–191. [36] Guo W, Xu W, Huvos AG, et al. Comparative frequency of bone [15] Dabak N, Cıraklı A, Gulman € B, et al. Distribution and evaluation sarcomas among different racial groups. Chin Med J. 1999; of bone and soft tissue tumors in the Middle Black Sea region. 112(12):1101–1104. Acta Orthop Traumatol Turc. 2014;48(1):17–24. [37] Polednak AP. Primary bone cancer incidence in black and white [16] DelaGarza-Montano P, Estrada-Villasenor E, Dominguez Rubio R, residents of New York State. Cancer. 1985;55(12):2883–2888. et al. Epidemiological aspects of osteosarcoma, giant cell tumor [38] Cleton-Jansen AM, van Beerendonk HM, Baelde HJ, et al. and chondrosarcoma musculoskeletal tumors–experience of the Estrogen signaling is active in cartilaginous tumors: implications national rehabilitation institute, Mexico city. Asian Pac J Cancer for antiestrogen therapy as treatment option of metastasized or Prev. 2015;16(15):6451–6455. irresectable chondrosarcoma. Clin Cancer Res. 2005;11(22):8028– [17] Aggerholm-Pedersen N, Maretty-Nielsen K, Baerentzen S, et al. Chondrosarcoma: the impact of comorbidity – 30 years of experi- [39] Meijer D, Gelderblom H, Karperien M, et al. Expression of aroma- ence from a population-based database including 199 consecu- tase and estrogen receptor alpha in chondrosarcoma, but no tive chondrosarcoma patients. ORR. 2019;11:109–116. beneficial effect of inhibiting estrogen signaling both in vitro and [18] Bergovec M, Kubat O, Smerdelj M, et al. Epidemiology of muscu- in vivo. Clin Sarcoma Res. 2011;1:5. loskeletal tumors in a national referral orthopedic department. A [40] Laitinen MK, Albergo JI, Stevenson JD, et al. Female gender in study of 3482 cases. Cancer Epidemiol. 2015;39(3):298–302. the hormonally active age group plays a major role in high- [19] Rivas-Berny C, M endez-Dom ınguez N, Alejos-Gomez  R. Prevalence grade chondrosarcoma survival. Acta Oncol. 2020;59(2):242– in Mexico of giant cell tumor, osteosarcoma and chondrosarcoma (2013–2017). Acta Ortop Mex. 2020;34(3):183–188. [41] Almen AFE, Widmark A, Olerud HM, et al. Radiologisk under- [20] Qureshi A, Ahmad Z, Azam M, et al. Epidemiological data for søkelser i norge per 2008. Trender i undersøkelsesfrekvens og common bone sarcomas. Asian Pac J Cancer Prev. 2010;11(2): stråledoser til befolkningen. Strålevern Rapport. 2010. 393–395. [42] Anfinsen KP, Devesa SS, Bray F, et al. Age-period-cohort analysis [21] Yucet € urk € G, Sabah D, Kec¸eci B, et al. Prevalence of bone and soft of primary bone cancer incidence rates in the United States tissue tumors. Acta Orthop Traumatol Turc. 2011;45(3):135–143. [22] Larsen IK. Cancer in Norway 2019- Cancer incidence, mortality, (1976–2005). Cancer Epidemiol Biomarkers Prev. 2011;20(8):1770– survival and prevalence in Norway. Oslo: Cancer Registry of [43] Deckers C, Schreuder BH, Hannink G, et al. Radiologic follow-up Norway 2020. 2020. of untreated enchondroma and atypical cartilaginous tumors in [23] Compton ML, Cates JMM. Evidence-based tumor staging of skel- etal chondrosarcoma. Am J Surg Pathol. 2020;44(1):111–119. the long bones. J Surg Oncol. 2016;114(8):987–991. [24] Giuffrida AY, Burgueno JE, Koniaris LG, et al. Chondrosarcoma [44] Deckers C, de Leijer EM, Flucke U, et al. Curettage and cryosur- in the United States (1973 to 2003): an analysis of 2890 cases gery for enchondroma and atypical cartilaginous tumors of the 8 J. THORKILDSEN AND T. Å. MYKLEBUST long bones: oncological results of a large series. J Surg Oncol. hundred and eight patients with a minimum follow-up of two 2021;123(8):1821–1827. years. J Bone Joint Surg Am. 2016;98(4):303–313. [45] Dierselhuis EF, Gerbers JG, Ploegmakers JJ, et al. Local treatment [46] Dierselhuis EF, Overbosch J, Kwee TC, et al. Radiofrequency ablation in with adjuvant therapy for Central atypical cartilaginous tumors in the treatment of atypical cartilaginous tumours in the long bones: les- the long bones: analysis of outcome and complications in one sons learned from our experience. Skeletal Radiol. 2019;48(6):881–887. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Oncologica Taylor & Francis

The national incidence of chondrosarcoma of bone; a review

Thorkildsen, Joachim; Myklebust, Tor Åge
Acta Oncologica , Volume 62 (2): 8 – Feb 1, 2023
Download PDF
8 pages

Loading next page...
 
/lp/taylor-francis/the-national-incidence-of-chondrosarcoma-of-bone-a-review-gG9nZq7gjC

References (47)

Publisher
Taylor & Francis
Copyright
© 2023 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
ISSN
1651-226X
eISSN
0284-186X
DOI
10.1080/0284186X.2023.2177975
Publisher site
See Article on Publisher Site

Abstract

ACTA ONCOLOGICA https://doi.org/10.1080/0284186X.2023.2177975 REVIEW a,b b,c Joachim Thorkildsen and Tor Åge Myklebust a b Division of Orthopaedic Surgery, Oslo University Hospital, Nydalen, Oslo, Norway; Department of Registration, Cancer Registry of Norway, Majorstuen, Oslo, Norway; Department of Research, Møre & Romsdal Hospital Trust, Ålesund, Norway ABSTRACT ARTICLE HISTORY Received 11 January 2023 Background: Chondrosarcoma (CS) epidemiology has been studied by a number of authors using Accepted 1 February 2023 national cancer registry cohorts. Many reports share the common findings of a slight increase in inci- dence, but not all. The patterns and causes for these changes are divergent while reflection concern- KEYWORDS ing methodological challenges are often missing. Chondrosarcoma; incidence; Method: We have performed a structured literature review to find national analyses of CS incidence review; epidemiology published from 2010 to 2020. We included eight studies of national incidence of CS, summarise their findings and patterns of change. We further discuss explanations given for these changes to better understand the real patterns and raise awareness in their interpretation. Results: Reported crude incidence ranges from 0.27 per million per year overall in Saudi Arabia to 5.4 in the Netherlands. Four studies from the USA, England, Switzerland and France report age standar- dised rates of 2.0–4.1 per million per year overall. While some countries report stable patterns, most report a slight increase. The Netherlands is the only country reporting a large increase, driven by a 10- fold increase in the incidence of ACT/grade 1 CS during the study period. We challenge the explana- tions given for this and suggest that this most likely is a result of variable interpretation and definition of CS at the lower levels of disease aggressiveness. This should raise awareness to possible over-treat- ment of CS in the Netherlands. Conclusion: The most likely national incidence of CS of bone is between 2–4 per million per year. Three modern reports present an incidence of 3.4–4.1 per million per year. Background Method The Norwegian Cancer Registry’s (NCR) main objective is to We have followed the PRISMA guidelines for review as far as establish knowledge and disseminate information that con- possible. Inclusion criteria are national cohorts with crude or age-standardised incidence rates (ASR), published in 2010– tributes to reducing cancer (https://www.kreftregisteret.no/ 2020. Exclusion criteria are institutional data sets, data for en). With this mandate, we published a national analysis of limited sites, subtypes or patient groups and prognostic incidence and prognostication for Chondrosarcoma (CS) of reports without incidence data. bone in Norway in 2019 [1]. The literature review and selection of articles are illus- A number of new articles with similar methodology, from trated in Figure 1. The main author completed a ‘Pubmed’ other countries, have been published [2,3] since then and a search for the terms ‘chondrosarcoma and incidence’ on the new edition of the WHO book has been released [4]. The 21.05.21. The search was limited to publication dates 2010– WHO state that the incidence of low-grade CS has increased 2020. This gave 445 results. The main author screened the over the past decade as a result of the improved sensitivity title and abstract or full text where necessary. of radiological scans and refers to a national epidemiological A total of 20 papers were identified as being epidemio- study from the Netherlands [5]. They do so without compara- logical studies including an assessment of the incidence of tive analysis, while the incidence reported in the Netherlands chondrosarcoma and were reviewed in full text. Assessment is by far the highest in the world. Furthermore, an inter- of these 20 papers led to the inclusion of eight national epi- national comparison of bone sarcoma epidemiology found demiological studies from the USA [7], England [7], The that CS was the bone sarcoma type with the most variability Netherlands [5], Norway [8], Switzerland [3], France [2], in reporting [6]. Taiwan[9] and Saudi Arabia[10]. A systematic analysis and comparison are therefore The other 12 were excluded. One study contained data for required to establish what the correct incidence of CS of certain grades of disease only[11] while another for specific bone is. Further, to explore potential causes of change and age range of patients[12]. Five studies were limited to regional variability in reporting. [10,13–17] and 4 institutional [18–21] data. One review paper CONTACT Joachim Thorkildsen jthork@ous-hf.no Division of Orthopaedic Surgery, Oslo University Hospital, PO Box 4956, Nydalen, 0424, Oslo, Norway This study was performed at Oslo University Hospital, Oslo. The interpretation and reporting of these data are the sole responsibility of the authors. 2023 Acta Oncologica Foundation 2 J. THORKILDSEN AND T. Å. MYKLEBUST Figure 1. Flow chart for literature review and inclusion of studies. was excluded though it contained data from multiple nations per year, they are converted to rates per million per year for through the International Agency for Research on Cancer [6]. ease of comparison. The overall crude incidence rate varied from 0.27 per million This is a database for records on cancer incidence based on in Saudi Arabia to 5.4 per million in the Netherlands. Four both regional and national registries, but it does not list the European reports in modern time periods find the crude inci- methodology at the individual registries. dence of 2.85, 4.2, 5.0 and 5.4 overall. Three studies report age We extracted data for crude and age-standardised inci- standardised rates with the European standard population from dence (ASR) rates by sex and overall, together with the 2.0–4.1 per million/year while three reports using the world standard population used for calculation from text, tables standard publish ASR of 1.2–3.4 per million per year. The high- and supplementary material published. We present only the est incidence of any type published in subgroups or time peri- extracted data in their original form. Table 1, summarises the ods was 8.78 in 2005–2013 for the Netherlands. Five nations features of the included studies. (USA, England, Netherlands, Norway and France) report an increase in the incidence of CS over the study period, whilst the Swiss study found no evidence of change. In Taiwan and Results Saudi Arabia, the patterns of change were not examined. The results from the included studies are summarised in Five studies demonstrated higher incidence (crude/ASR) Table 2. Where ASR’s are presented per 100,000 inhabitants values for men than for women, while in the USA the rate ACTA ONCOLOGICA 3 Table 1. Summary of included studies ASR ¼ Age standardised rate. Surname main Incidence Author / year Inclusion Nation of type Population of publication period origin Register name Register type reported standard used Arnfinsen 2011 1976–2005 USA Surveillance, Epidemiology, and Combination of 9 regional ASR World End Results program (SEER) cancer registries Whelan 2012 1998–2007 England National Cancer Data Combination of 8 regional ASR European Repository (NCDR) cancer registries Hung 2014 2003–2010 Taiwan Taiwan Cancer Registry (TCR) National cancer registry ASR World van Praag 2018 1989–2013 Netherlands Netherlands Cancer Registry National cancer registry Crude None (NCR-IKNL) Thorkildsen 2019 1990–2013 Norway Cancer Registry of National cancer registry Crude None Norway (CRN) Kollar 2019 1996–2015 Switzerland National Institute for Cancer Combination of Swiss Crude ASR European Epidemiology and Cantonal Cancer Registration (NICER) Registries Amadeo 2020 2000–2013 France French Network of cancer Combination of 19 regional Crude ASR European, Registries (FRANCIM) cancer registries world, US Aljuhani 2020 2013-2017 Saudi Arabia Saudi Cancer Registry (SCR) National cancer registry Crude None Table 2. Summary of results from included studies. Surname main Crude Highest Time Explanation Author rate ASR Incidence (type) trend proposed Arnfinsen 2011 Na 2.7 men 2.7 (ASRw) Increase Exogenous estrogen 2.7 women Whelan 2012 Na 2.6 men 2.0 (ASRe) Increase, then Change in reporting 1.7 women stabilisation 2.0 total Hung 2014 Na 1.34 men 1.2 (ASRw) Not examined Na 1.02 women 1.2 total van Praag 2018 5,4 overall Na 8.78 (Crude) Increase Increase in ACT due to increased imaging and aging population Thorkildsen 2019 3.04 men Na 3.45 (Crude) Increase None given 2.67 women 2.85 overall Kollar 2019 5,0 men 4.3 men 3.4 (ASRe) No change Na 3.5 women 3.0 women 4.2 total 3.4 total Amadeo 2020 5.0 men 4.1 men 4.1 (ASRe) Increase Exogenous estrogen 4.0 women 3.6 women 5.0 total 4.1 total Aljuhani 2020 0.27 total Na 0.27 (Crude) Not examined Na a b Arnfinsen reports over period 1976–2005, but in three time periods. The most recent time period features her is 1996–2005. ASR given for European, world and US reference. European is used here. ASRw: Age standardised rate by world standard population, ASRe: Age standardised rate by European standard popu- lation. Na:not available. was similar between the sexes. Three studies included data The Dutch present a crude overall incidence of 2.88 per on incidence over time, by sex. In England, Whelan et al. million per year for 1989–96 increasing to 8.78 for 2005–13. showed an increased ASR for both sexes (37% for men and The increase is driven by a rise in the incidence of 40% for women), mostly occurring during the 1980s with sta- ACT/grade 1 CS (2013 WHO definition), with an increase bilization thereafter. In the USA however, there was no from 1.2 per million in 1989–96 to 6.63 in 2005–13. change in the ASR incidence among men, but an increase Extracting data from their Figure 2 shows a crude incidence for women through the study period of 1976–2005. In for ACT/grade 1 CS in 2011 and 2012 of 9.8 per million per Norway, there was an increase in crude incidence for both year. Van Praag et al. also show a doubling of the incidence sexes (23% for men and 82% for women). The increase for of grade 2 disease from 0.68 per million per year1989–96 to women was such that the rate was equal, compared to men 1.38 per million per year in 2005–13. Grade 3 disease in the final segment of the study (2009–2013). increases from 0.27 per million per year in 1989–96 to 0.41 The studies from Norway and the Netherlands both have per million per year in 1997–04, but then remains stable at more detailed data at the subtype level allowing us to look 0.43 per million per year in 2005–2013. These last two peri- beyond sex. The increase in Norway is driven by a rise in the ods show an incidence of grade 3 CS remarkably similar to incidence of the central CS subtype, which increases with the Norwegian numbers. age. The increase was evident for grade 1 and 2 diseases For the most modern time period, excluding the while grade 3 disease remained stable. The increase was Netherlands, the incidence of CS was 3.4 per million per year largest for grade 2 disease. for Switzerland (ASR-europe), 3.45 for Norway (Crude) and 4 J. THORKILDSEN AND T. Å. MYKLEBUST Figure 2. Incidence of ACT/CS grade 1 in the Netherlands, between 1889 and 2013, number of MRI examinations over time in the Netherlands. Adapted from van Praag et al. with addition of number of MRI examinations in Norway. 4.1 for France (ASR-Europe). Similar findings, from similar Five authors present rates by ASR, but another methodical periods of time and countries, support the validity of the dimension is that ASR can be calculated with a range of dif- findings. ferent standard populations as presented in Table 1. Different age groups are weighted differently in different ref- erence populations. As such, this can give significant differ- Discussion ences. Two authors use the world standard, two use the European standard and one uses multiple standards. This has In a global assessment of bone sarcoma incidence from 68 been discussed in the annual Norwegian Cancer Registry countries reporting to the International Agency for Reporting report of 2019 in relation to their usage of the Norwegian of Cancer (IARC) for 2003–2007, only small changes in inci- Standard [22]. As illustrated by Figure 3, the Norwegian dence over time are noted for CS [6]. Although increases standard weights the elderly population significantly more have been reported for both men and women in the than the world standard. The overall national cancer ASR per Netherlands [5] and Norway [1], the IARC report shows 100,000 per year for all sites in Norway is 700 for men by equivalent decreases in CS rates among men in Sweden and the Norwegian standard while only 360 by the world stand- women in Ireland and Japan.[6] ard. For ‘Bone’ location however it is 1.2 for the Norwegian Valery et al. also describe the largest variability of report- standard while 1.0 by the world standard. Amadeo et al. ing for CS, compared to Ewing and Osteosarcoma. CS repre- illustrate this by presenting their numbers with both. The sents <10% of reported bone sarcomas in India and Saudi French national incidence is 20% higher if calculated by the Arabia, while >45% in Finland, Slovenia and the Netherlands. European standard (4.1 per million per year) as compared to They also state that although CS incidence rose with age in the world standard (3.4 per million per year) [2]. We have all countries, a peak was seen earlier and more pronounced attempted to list these different standards and methods in countries with higher incidence. Varied patterns of report- clearly in both text and tables to allow for comparison and ing will naturally lead to variability in estimated incidence. raise awareness. Although an important point of accuracy, There are some important methodical challenges to an this difference can not explain the size of the differences accurate comparison of incidence. First, key publications seen between the Dutch numbers at the upper end and the from Norway and the Netherlands include crude incidence Saudi Arabian at the lower end. only. These are figures for the number of cases of CS over a With regards to methods of reporting, the articles period of time divided by the population at risk in the nation included also represent different types of registries. Norway, in the same study time without adjustment. Both these pub- the Netherlands and Taiwan have a single national registry. lications of national cohorts present crude incidence despite both national registries from which they are drawn routinely The UK, France and Switzerland have a national registry that using ASR to present incidence. ASR is observed age-specific combines data from regional registries with presumed rates applied to a standard population and adjusted by a national complete coverage. weight depending on the composition of the standard popu- The USA has a registry that also combines data from lation used. They thereby give a rate adjusted for age, allow- regional registries, but which covers 30% of the US popula- ing comparison between groups and over time. tion. They then extrapolate data based on this. A recent ACTA ONCOLOGICA 5 Figure 3. Comparison of population weights for Norwegian and world standard reference populations from Cancer Registry of Norway. assessment of evidence-based CS staging from the US noted this basis, it is likely that differences in access to-, and that up to 40% of skeletal CS entries in the SEER database resources of the health care system correlate to the amount lacked essential variable information [23]. In SEER publica- of CS discovered. In particular, it is likely as van Praag et al. tions, up to 10% of cases included are termed myxoid chon- argue, that the frequency of use of diagnostic radiology serv- drosarcoma [24,25]This is not a separate subtype in the WHO ices would likely correlate to the volume of CS discovered. categorization of chondrogenic sarcoma, but rather myxoid The worldbank publishes statistics for current health change is a part of pathological grading. It is possible that expenditure per capita in US$(https://data.worldbank.org/). these cases in fact represent extraskeletal myxoid sarcoma The values of US$per capita spending in 2019 range from which is a soft tissue sarcoma with a misleading name since $10 623 for the USA at the upper end, to $1316 for Saudi these are not organised under chondrogenic differentiation, Arabia at the lower end. Taiwan is not listed in the statistics. but rather uncertain differentiation by the WHO. There is a These numbers confirm that there is a substantial difference publication of an exceedingly rare CS subtype termed myx- in the health care systems of the countries included and this oid CS or Chordoid sarcoma [26], but it seems unlikely that will likely represent a certain difference in the use of diag- this represents 10% of a national cohort not reported by nostic radiology and thereby CS diagnosis, but also in meth- other countries. ods and resources used in reporting cancer. Norway has mandatory reporting of CS to the register by A further unique feature of CS is its well-established law for all personnel involved in cancer care since 1952. inter-and intra-observer variability in assessment by radiol- The Netherlands also has mandatory reporting by law. The ogists, pathologists and clinicians [31–33]. This is most evi- Cancer Control Act of 2003 in Taiwan also obligates all hos- dent at the lower level of disease aggressiveness, in pitals with greater than 50-bed capacity, involved in cancer distinguishing between benign enchondroma and grade 1 care, to participate in reporting all newly diagnosed malig- CS or ACT. At the same time, central chondroid lesions of nant neoplasms [27]They estimate that 80% of cancer the skeleton appear in approximately 2–3% of routinely per- patients are diagnosed and treated in such units. The formed radiological examinations of the knee and shoulder Norwegian register has published rates of completeness [34,35]. There is no uniform accepted practice for the assess- [28], but similarly does Taiwan [27]. A comparative analysis ment or follow-up of these lesions. It is methodically likely of the national cancer registries of the Nordic countries of that this spectrum of disease is central to any difference in reporting. Norway, Sweden, Finland, Denmark and Iceland concluded The lower numbers in Taiwan and Saudi Arabia can of that although these were more similar than any other col- lection of data from five different countries there were sig- course also be an expression of ethnic difference in inci- nificant differences in registration, classification and dence as suggested by a comparative analysis of Japanese, inclusion [29]. Chinese, and US bone sarcoma cases [36]. An epidemio- A key feature of CS epidemiology is that the overall prog- logical study from New York in the USA found that the nosis for CS is better and that tumour growth is more indo- reported incidence for CS was lower for the ethnic black lent than other primary bone sarcomas. It has been shown population as compared to the ethnic white population [37]. that 6% of diagnosed CS were asymptomatic [30] while in They argued that this partially can be explained by the black the more recent Norwegian study, as much as 15% of CS population having shorter life expectancy and therefore was retrospectively categorised as ‘incidental findings’ [1]. On lower at-risk exposure in the elderly age groups where CS is 6 J. THORKILDSEN AND T. Å. MYKLEBUST more common. Ethnic difference is however not a finding The Dutch increase was driven by a massive rise in found for CS by an international comparative analysis of ACT/CS grade 1, not experienced by comparable countries. bone cancer epidemiology [6]. Histological grading has not formally changed during the Two articles quote exogenous estrogens as a likely associ- period. At the same time this is the entity most open to ation with increased levels of CS. The study of CS and estro- interobserver variation. The natural explanation for their gens is however divergent. Although estrogen signaling has increase is one of variable definitions and a more careful been shown to be active in cartilaginous tumours [38], the practice. It seems likely that the Netherlands have a lower study of inhibiting estrogen in vitro and in vivo showed no level of interpretation and intervention for central chondroid beneficial effect [39]. In addition, Laitinen et al. showed that lesions compared to other countries. In the assessment of a being hormonally active appears to give survival advantages central chondroid lesion they are more likely to conclude in women with CS [40]. with the likelihood of ACT/grade 1 CS than enchondroma, The two most similar reports come from Norway and the compared to other nations. The scale of this appears to be Netherlands. Both studies provide subtype-specific crude quite large. We cannot presume to know the cause or reason incidence data over a similar modern time period from well- for this practice. Whatever the reason, it most probably rep- functioning and well-resourced national registries in similar resents a significant level of over-treatment. This is sup- societies and populations. All the same, they report vastly ported by the publication of safe observation without different results, driven by different subgroups of disease. treatment of extremity grade 1 CS/ACT [43] in the Van Praag et al. offer an explanation of the increased use of Netherlands. There is no cause in the literature to suspect diagnostic images and an aging population. Both of these under-treatment of these entities. Even with minimally inva- explanations can be further examined. sive procedures, CS management is associated with consider- In Norway, the Norwegian Radiation and Nuclear Safety able morbidity [44–46]. Authority provide interim reports on the use of radiological examinations at a national and international level[41]. They demonstrate in their reports from 2002 and 2008 that the Conclusion rates of radiological investigations performed was slightly Removing the outliers it appears that the incidence of CS is higher in Norway compared to the Netherlands in one report likely to be in the range of 2-4 cases per million inhabitants and conversely in the other. The number of MRI performed per year. Two reports using similar population standards pre- per 1000 inhabitants in Norway per year was 61 in 2002 and sent ASR rates of 3.4 and 4.1 per million per year. A majority 126 in 2008 [43]. These numbers can then be entered into van Praag’s adapted figure (Figure 2) as red stars. Provided of countries report a slight increase. This may in part be that a proportion of CS are incidental findings, an increase in explained by increased use of diagnostic imaging. The level use of MRI can support a limited increase in CS incidence, of CS reported by the Netherlands driven by increasing but not of the same magnitude as that reported from the ACT/grade 1 CS is uniquely high. The most likely explanation Netherlands. Furthermore, with similar use of MRI in Norway is one of variable definitions of this lesion and possible over- and the Netherlands, the change in CS incidence should also treatment. be similar, which it is not. This lends supports to the notion that the same lesions are probably interpreted, managed Disclosure statement and reported differently in the two nations. It is known that the incidence of CS rises with age [1,7,42] No potential conflict of interest was reported by the author(s). and as such the notion that an aging population can explain an increase in CS incidence is logical. Population data for the Funding Netherlands is readily available from the Statistics Netherlands (http://www.cbs.nl/en-gb/figures/detail/37556eng). There was None to disclose an increase of 33% in the proportion of the population older than 65 years of age from 1990–2010, or 632,000 people in ORCID total. These numbers are far too small in absolute terms to explain a 10-fold rise in incidence of grade 1 CS which is a Tor Åge Myklebust http://orcid.org/0000-0003-4645-1635 rare tumor even when using the reported Dutch estimates. This can, only to a limited degree, explain a rise in the inci- Data availability statement dence of CS. Further, if they used age standardised rates, rather than crude rates, this change in population would All dara are available from the main author upon reasonable request. already be adjusted for in their rate. The Netherlands is a leading voice in CS research and a References defining force for CS terminology through the WHO. The findings and explanations of van Praag et al. are reproduced [1] Thorkildsen J, Taksdal I, Bjerkehagen B, et al. Chondrosarcoma th in the 5 edition WHO book [4], but in our opinion with the in Norway 1990–2013; an epidemiological and prognostic insufficient investigation. The estimated numbers from the observational study of a complete national cohort. Acta Oncol. Netherlands need further explanation. 2019;58(3):1–10. ACTA ONCOLOGICA 7 [2] Amadeo B, Penel N, Coindre JM, et al. Incidence and time trends from the SEER database. J Bone Joint Surg Am. 2009;91(5): of sarcoma (2000–2013): results from the French network of can- 1063–1072. cer registries (FRANCIM). BMC Cancer. 2020;20(1):190. [25] Amer KM, Munn M, Congiusta D, et al. Survival and prognosis of [3] Kollar A, Rothermundt C, Klenke F, et al. Incidence, mortality, and chondrosarcoma subtypes: SEER database analysis. J Orthop Res. survival trends of soft tissue and bone sarcoma in Switzerland 2020;38(2):311–319. between 1996 and 2015. Cancer Epidemiol. 2019;63:101596. [26] Kilpatrick SE, Inwards CY, Fletcher CD, et al. Myxoid chondrosar- [4] WHO Classification of Tumours Editorial Board. Soft tissue and coma (chordoid sarcoma) of bone: a report of two cases and bone tumours. 5th ed. Board WCE, editor. Lyon (France): review of the literature. Cancer. 1997;79(10):1903–1910. International Agency for Research on Cancer; 2020. [27] Chiang CJ, You SL, Chen CJ, et al. Quality assessment and [5] van Praag VM, Rueten-Budde AJ, Ho V, et al. Incidence, outcomes improvement of nationwide cancer registration system in Taiwan: and prognostic factors during 25 years of treatment of chondro- a review. Jpn J Clin Oncol. 2015;45(3):291–296. sarcomas. Surg Oncol. 2018;27(3):402–408. [28] Larsen IK, Smastuen M, Johannesen TB, et al. Data quality at the [6] Valery PC, Laversanne M, Bray F. Bone cancer incidence by mor- cancer registry of Norway: an overview of comparability, com- phological subtype: a global assessment. Cancer Causes Control. pleteness, validity and timeliness. Eur J Cancer. 2009;45(7):1218– 2015;26(8):1127–1139. [7] Whelan J, McTiernan A, Cooper N, et al. Incidence and survival of [29] Pukkala E, Engholm G, Højsgaard Schmidt LK, et al. Nordic cancer malignant bone sarcomas in England 1979–2007. Int J Cancer. registries - an overview of their procedures and data comparabil- 2012;131(4):E508–E517. ity. Acta Oncol. 2018;57(4):440–455. [8] Thorkildsen J, Taksdal I, Bjerkehagen B, et al. Chondrosarcoma in [30] BjoR € nsson J, McLeod RA, Unni KK, et al. Primary chondrosar- Norway 1990–2013; an epidemiological and prognostic observa- coma of long bones and limb girdles. Cancer. 1998;83(10):2105– tional study of a complete national cohort. Acta Oncol. 2019; 58(3):273–282. [31] Eefting D, Schrage YM, Geirnaerdt MJ, et al. Assessment of inter- [9] Hung GY, Horng JL, Yen HJ, et al. Incidence patterns of primary observer variability and histologic parameters to improve reliabil- bone cancer in Taiwan (2003–2010): a population-based study. ity in classification and grading of Central cartilaginous tumors. Ann Surg Oncol. 2014;21(8):2490–2498. Am J Surg Pathol. 2009;33(1):50–57. [10] Aljuhani WS, Alanazi AM, Alghafees MA. Primary bone sarcomas [32] Skeletal Lesions Interobserver Correlation among Expert in KSA: a Saudi tumor registry review. J Taibah Univ Medical Sci. Diagnosticians (SLICED) Study Group Reliability of histopathologic 2021;16(1):77–85. and radiologic grading of cartilaginous neoplasms in long bones. [11] Goedhart LM, Ho VKY, Dijkstra PDS, et al. Bone sarcoma incidence J Bone Joint Surg Am. 2007;89:2113–2123. in The Netherlands. Cancer Epidemiol. 2019;60:31–38. [33] Zamora T, Urrutia J, Schweitzer D, et al. Do orthopaedic oncolo- [12] Eyre R, Feltbower RG, James PW, et al. The epidemiology of bone gists agree on the diagnosis and treatment of cartilage tumors of cancer in 0–39 year olds in Northern England, 1981 - 2002. BMC the appendicular skeleton? Clin Orthop Relat Res. 2017;475(9): Cancer. 2010;10(1):357. 2176–2186. [13] Ducimetiere F, Lurkin A, Ranchere-Vince D[, et al. Incidence rate, [34] Stomp W, Reijnierse M, Kloppenburg M, et al. Prevalence of cartil- epidemiology of sarcoma and molecular biology. Preliminary aginous tumours as an incidental finding on MRI of the knee. Eur results from EMS study in the Rhone-Alpes region. Bull Cancer. Radiol. 2015;25(12):3480–3487. 2010;97:629–641. [35] Hong ED, Carrino JA, Weber KL, et al. Prevalence of shoulder [14] Solooki S, Vosoughi AR, Masoomi V. Epidemiology of musculo- enchondromas on routine MR imaging. Clin Imaging. 2011;35(5): skeletal tumors in shiraz, South of Iran. Indian J Med Paediatr 378–384. Oncol. 2011;32(4):187–191. [36] Guo W, Xu W, Huvos AG, et al. Comparative frequency of bone [15] Dabak N, Cıraklı A, Gulman € B, et al. Distribution and evaluation sarcomas among different racial groups. Chin Med J. 1999; of bone and soft tissue tumors in the Middle Black Sea region. 112(12):1101–1104. Acta Orthop Traumatol Turc. 2014;48(1):17–24. [37] Polednak AP. Primary bone cancer incidence in black and white [16] DelaGarza-Montano P, Estrada-Villasenor E, Dominguez Rubio R, residents of New York State. Cancer. 1985;55(12):2883–2888. et al. Epidemiological aspects of osteosarcoma, giant cell tumor [38] Cleton-Jansen AM, van Beerendonk HM, Baelde HJ, et al. and chondrosarcoma musculoskeletal tumors–experience of the Estrogen signaling is active in cartilaginous tumors: implications national rehabilitation institute, Mexico city. Asian Pac J Cancer for antiestrogen therapy as treatment option of metastasized or Prev. 2015;16(15):6451–6455. irresectable chondrosarcoma. Clin Cancer Res. 2005;11(22):8028– [17] Aggerholm-Pedersen N, Maretty-Nielsen K, Baerentzen S, et al. Chondrosarcoma: the impact of comorbidity – 30 years of experi- [39] Meijer D, Gelderblom H, Karperien M, et al. Expression of aroma- ence from a population-based database including 199 consecu- tase and estrogen receptor alpha in chondrosarcoma, but no tive chondrosarcoma patients. ORR. 2019;11:109–116. beneficial effect of inhibiting estrogen signaling both in vitro and [18] Bergovec M, Kubat O, Smerdelj M, et al. Epidemiology of muscu- in vivo. Clin Sarcoma Res. 2011;1:5. loskeletal tumors in a national referral orthopedic department. A [40] Laitinen MK, Albergo JI, Stevenson JD, et al. Female gender in study of 3482 cases. Cancer Epidemiol. 2015;39(3):298–302. the hormonally active age group plays a major role in high- [19] Rivas-Berny C, M endez-Dom ınguez N, Alejos-Gomez  R. Prevalence grade chondrosarcoma survival. Acta Oncol. 2020;59(2):242– in Mexico of giant cell tumor, osteosarcoma and chondrosarcoma (2013–2017). Acta Ortop Mex. 2020;34(3):183–188. [41] Almen AFE, Widmark A, Olerud HM, et al. Radiologisk under- [20] Qureshi A, Ahmad Z, Azam M, et al. Epidemiological data for søkelser i norge per 2008. Trender i undersøkelsesfrekvens og common bone sarcomas. Asian Pac J Cancer Prev. 2010;11(2): stråledoser til befolkningen. Strålevern Rapport. 2010. 393–395. [42] Anfinsen KP, Devesa SS, Bray F, et al. Age-period-cohort analysis [21] Yucet € urk € G, Sabah D, Kec¸eci B, et al. Prevalence of bone and soft of primary bone cancer incidence rates in the United States tissue tumors. Acta Orthop Traumatol Turc. 2011;45(3):135–143. [22] Larsen IK. Cancer in Norway 2019- Cancer incidence, mortality, (1976–2005). Cancer Epidemiol Biomarkers Prev. 2011;20(8):1770– survival and prevalence in Norway. Oslo: Cancer Registry of [43] Deckers C, Schreuder BH, Hannink G, et al. Radiologic follow-up Norway 2020. 2020. of untreated enchondroma and atypical cartilaginous tumors in [23] Compton ML, Cates JMM. Evidence-based tumor staging of skel- etal chondrosarcoma. Am J Surg Pathol. 2020;44(1):111–119. the long bones. J Surg Oncol. 2016;114(8):987–991. [24] Giuffrida AY, Burgueno JE, Koniaris LG, et al. Chondrosarcoma [44] Deckers C, de Leijer EM, Flucke U, et al. Curettage and cryosur- in the United States (1973 to 2003): an analysis of 2890 cases gery for enchondroma and atypical cartilaginous tumors of the 8 J. THORKILDSEN AND T. Å. MYKLEBUST long bones: oncological results of a large series. J Surg Oncol. hundred and eight patients with a minimum follow-up of two 2021;123(8):1821–1827. years. J Bone Joint Surg Am. 2016;98(4):303–313. [45] Dierselhuis EF, Gerbers JG, Ploegmakers JJ, et al. Local treatment [46] Dierselhuis EF, Overbosch J, Kwee TC, et al. Radiofrequency ablation in with adjuvant therapy for Central atypical cartilaginous tumors in the treatment of atypical cartilaginous tumours in the long bones: les- the long bones: analysis of outcome and complications in one sons learned from our experience. Skeletal Radiol. 2019;48(6):881–887.

Journal

Acta Oncologica – Taylor & Francis

Published: Feb 1, 2023

Keywords: Chondrosarcoma; incidence; review; epidemiology

There are no references for this article.