Dermatofibrosarcoma protuberans (DFSP) is

Transcription

1 Incidence and Survival of Primary Dermatofibrosarcoma Protuberans in the United States Kathryn L. Kreicher, BA,* David E. Kurlander, MD, Haley R. Gittleman, MS, Jill S. Barnholtz-Sloan, PhD, and Jeremy S. Bordeaux, MD, MPH* BACKGROUND Dermatofibrosarcoma protuberans (DFSP) is a rare cutaneous sarcoma for which data on risk factors, incidence, and survival are limited. OBJECTIVE The authors sought to establish a comprehensive report on the incidence of and survival from primary DFSP. METHODS The authors used data from the 18 registries of the Surveillance, Epidemiology, and End Results Program from 2000 to RESULTS Overall incidence was 4.1 per million person-years and steady over the decade. Trunk was the most common anatomic site except in older men. Incidence among women was 1.14 times higher than men (95% confidence interval [CI] of rate ratio: ). Incidence among blacks was almost 2 times the rate among whites (95% CI of rate ratio: ). Ten-year relative survival of DFSP was 99.1% (95% CI: ). Increased age, male sex, black race, and anatomic location of the limbs and head as compared with the trunk were associated with higher all-cause mortality. CONCLUSION This is the largest population-based study of DFSP derived from a cohort of almost 7,000 patients. The epidemiologic profile of DFSP differs from most skin cancers. Incidence is stable and highest among women and blacks. Worse survival is associated with increased age, male sex, black race, and anatomic location of the limbs and head. The authors have indicated no significant interest with commercial supporters. Dermatofibrosarcoma protuberans (DFSP) is a rare cutaneous sarcoma of intermediate malignancy. More than 90% of DFSP tumors are characterized by a t(17;22) translocation that produces a platelet-derived growth factor beta (PDGF-b)/collagen Type 1A1 fusion gene, 1 3 leading to the constitutive production of collagen. Dermatofibrosarcoma protuberans is notable for its progressive growth and high local recurrence after surgical resection 4 6 but for low metastatic potential. Only 2 US population-based studies have been published on DFSP, which used data from 9 registries of the Surveillance, Epidemiology, and End Results (SEER, SEER-9) Program of the US National Cancer Institute (NCI). 7,8 The most recent estimates of overall incidence of DFSP in the US are between 4.2 and 5.0 cases per million. 7,8 Data on DFSP survival are very limited. In this study, the authors establish an updated and comprehensive report on incidence and survival of DFSP using all 18 registries from the SEER database through Materials and Methods Institutional Review Board approval was waived by University Hospitals Case Medical Center, Cleveland, OH. Our incidence and survival data were derived from the 18 registries of the SEER program of the NCI Departments of *Dermatology, and Plastic Surgery, University Hospitals Case Medical Center, Case Western Reserve University, Cleveland, Ohio; Department of Epidemiology and Biostatistics, Case Comprehensive Cancer Center, Cleveland, Ohio 2015 by the American Society for Dermatologic Surgery, Inc. Published by Wolters Kluwer Health, Inc. All rights reserved. ISSN: Dermatol Surg 2016;42:S24 S31 DOI: /DSS S24

2 KREICHER ET AL from 2000 to 2010, referred to as SEER SEER-18 was established in 2000 and now accounts for 27.8% of the total US population. SEER registries collect data on patient demographics, tumor characteristics (primary site, morphology, and state at diagnosis), treatment, and vital status. Patients given the diagnosis of DFSP as their first cancer, who lived at least 2 months beyond their diagnosis date, were included in this study. Dermatofibrosarcoma protuberans diagnosis was defined as code 8832/3 in the International Classification of Diseases for Oncology, Third Edition. 10 Autopsy cases and patients only given a diagnosis by death certificate were excluded from analysis. Data for the pigmented form of DFSP (Bednar tumor), coded as 8833/3, was examined separately. Descriptive analysis of the SEER-18 DFSP population was performed. Cases were grouped into 2 sexes: male and female; 5 age-at-diagnosis groups: #19 years, 20 to 39 years, 40 to 59 years, 60 to 79 years, $80 years; 4 racial groups: white, black, other (Native American/ Alaska Native, Native Hawaiian/Pacific Islander), and unknown; and 6 anatomic tumor sites: trunk, upper limb, lower limb, head, genitals, and other. Cases were further stratified by anatomic site according to age at diagnosis. Observed and relative 5- and 10-year survival were measured with SEER*Stat using the Kaplan Meier method. Relative survival is defined as the ratio of observed survival in a cohort of patients with DFSP to the proportion of expected survival in a cohort of cancer-free individuals adjusted for race, sex, age, and date at which the age was coded. Cox regression models in SEER*Stat were used to calculate hazard ratios (HRs) to determine if age, race, sex, anatomic site, and geographical location were associated with survival among patients with DFSP. Hazard ratios were also calculated using multivariate analysis controlling for age, race, sex, and anatomic site. Reference categories for age, race, sex, anatomic site, and geographical location were 19 years and younger, white, female, trunk, and San Francisco Oakland SMSA, respectively. The authors used the cause-specific death classification in SEER*Stat to analyze patients whose death was attributable to DFSP. Descriptive analysis of this population subset was performed. Dermatofibrosarcoma protuberans cause-specific death cases were grouped into 2 sexes: male and female; 4 age-atdiagnosis groups: #39 years, 40 to 59 years, 60 to 79 years, $80 years; 2 racial groups: white and black; and 6 anatomic sites: trunk, upper limb, lower limb, head, genitals, and other. The data were analyzed using SEER*Stat Incidence rates were calculated using SEER*Stat and expressed as cases per million person-years. Overall incidence from 2000 to 2010 and annual incidence rates were calculated for the entire DFSP population. Overall incidence rates and incidence rate ratios were calculated for each race, sex, and registry. Unknown race was excluded from race analyses because of small sample size. Age-adjusted incidence (age adjusted to the 2000 US standard population) according to the calendar year at diagnosis was calculated for males, females, whites, and blacks separately. SEER*Stat Joinpoint regression analysis was used to calculate annual percent change (APC) in incidence rates over time for all population subsets (whole population, male, female, white, black). Agespecific incidences for the entire DFSP population and by sex and race were also calculated. Results A total of 7,250 cases of DFSP were identified between 2000 and Four hundred thirty-three cases were excluded because they were secondary to other primary cancers. Dermatofibrosarcoma protuberans accounted for 0.1% of all cancers in SEER. The major characteristics of the cases are reported in Table 1. Of 6,817 primary DFSP cases reported from 2000 to 2010, 53.1% of cases were female and 46.9% were male, 70.9% of cases were whites, and 52% of cases were reported in people older than 40 years (Table 1). For both sexes, the age group with the most reported cases was 20 to 39 years (42.0%). Of all cases, 41.7% were on the trunk, 21.2% on the upper limb, 20.8% on the lower limb, 12.9% on the head, 1.0% on the genitals, and 2.4% other (Table 1). In females, trunk was the most common anatomic location for all age 42:1S:JANUARY SUPPLEMENT 2016 S25

3 EPIDEMIOLOGY OF DFSP TABLE 1. Characteristics of the 6,817 Patients With First Primary DFSP as Reported to the Surveillance, Epidemiology, and End Results Program (SEER ) Characteristics Males, N = 3,197 (46.9%), Females, N = 3,620 (53.1%), Total, N = 6,817, Age groups, yrs # (6.1) 213 (5.9) 409 (6.0) ,335 (41.8) 1,528 (42.2) 2,863 (42.0) ,196 (37.4) 1,378 (38.1) 2,574 (37.7) (12.6) 427 (11.8) 829 (12.2) $80 68 (2.1) 74 (2.0) 142 (2.1) Race White 2,295 (71.8) 2,538 (70.1) 4,833 (70.9) Black 560 (17.5) 704 (19.4) 1,264 (18.5) Other 240 (7.5) 249 (6.9) 489 (7.2) Unknown 102 (3.2) 129 (3.6) 231 (3.4) Anatomic site Trunk 1,220 (38.2) 1,621 (44.8) 2,841 (41.7) Upper limb 734 (22.9) 708 (19.6) 1,442 (21.2) Lower limb 652 (20.4) 768 (21.2) 1,420 (20.8) Head 499 (15.6) 381 (10.5) 880 (12.9) Genitals 9 (0.3) 62 (1.7) 71 (1.0) Other 83 (2.6) 80 (2.2) 163 (2.4) Anatomic site by age at diagnosis #19 yrs Trunk 71 (36.2) 85 (39.9) 156 (38.1) Upper limb 36 (18.4) 50 (23.5) 86 (21.0) Lower limb 55 (28.1) 50 (23.5) 105 (25.7) Head 28 (14.3) 25 (11.7) 53 (13.0) Genitals 1 (0.5) 1 (0.5) 2 (0.5) Other 5 (2.5) 2 (0.9) 7 (1.7) yrs Trunk 498 (37.3) 685 (44.8) 1,183 (41.3) Upper limb 319 (23.9) 308 (20.2) 627 (21.9) Lower limb 296 (22.2) 310 (20.3) 606 (21.2) Head 182 (13.6) 163 (10.7) 345 (12.0) Genitals 3 (0.2) 28 (1.8) 31 (1.1) Other 37 (2.8) 34 (2.2) 71 (2.5) yrs Trunk 483 (40.4) 624 (45.3) Upper limb 287 (24.0) 265 (19.2) 552 (21.4) Lower limb 220 (18.4) 294 (21.4) 514 (20.0) Head 173 (14.5) 138 (10.0) 311 (12.1) Genitals 4 (0.3) 25 (1.8) 29 (1.1) Other 29 (2.4) 32 (2.3) 61 (2.4) yrs Trunk 155 (38.6) 195 (45.6) 350 (42.2) Upper limb 82 (20.4) 71 (16.6) 153 (18.4) Lower limb 70 (17.4) 99 (23.2) 169 (20.4) Head 87 (21.6) 43 (10.1) 130 (15.7) Genitals 1 (0.3) 8 (1.9) 9 (1.1) Other 7 (1.7) 11 (2.6) 18 (2.2) S26 DERMATOLOGIC SURGERY

4 KREICHER ET AL TABLE 1. (Continued) Characteristics Males, N = 3,197 (46.9%), Females, N = 3,620 (53.1%), Total, N = 6,817, $80 yrs Trunk 13 (19.1) 32 (43.2) 45 (31.7) Upper limb 10 (14.7) 14 (18.9) 24 (16.9) Lower limb 11 (16.2) 15 (20.3) 26 (18.3) Head 29 (42.6) 12 (16.2) 41 (28.9) Genitals 0 (0.0) 0 (0.0) 0 (0.0) Other 5 (7.4) 1 (1.4) 6 (4.2) groups (Table 1). In males, the proportion of cases located at the head increased from 14.1% in males younger than 60 years to 21.6% in males aged 60 to 79 years and 42.6% of cases in males 80 years and older (Table 1). Staging data for DFSP are limited to an LRD staging system: localized (confined entirely to the organ of origin), regional (extends beyond the limits of the organ or origin directly into surrounding organs or tissues or into regional lymph nodes), or distant (spread to parts of body remote from the primary tumor either by direct extension or discontinuous metastasis). A total of 3,792 cases were localized, 2,314 were regional, 37 were distant, and 674 were unstaged. Of cases reported as distant, 22 were male and 15 female; 25 occurred in whites, 10 in blacks, and 2 in other races. Histological grade was not reported for the majority of cases. There were 128 cases of the rare pigmented variant of DFSP (Bednar tumor), of which 44.5% were male and 55.5% female. Fiftyseven percent of Bednar tumors occurred in whites, 28.9% blacks, 10.2% other, and 3.9% unknown. Anatomically, 33.6% were on the trunk, 28.1% on the upper limb, 28.9% on the lower limb, 8.6% on the head, and 0.8% other. Overall annual incidence rate of DFSP from 2000 to 2010 was 4.1 persons per million person-years. Incidence among women from 2000 to 2010 was 1.14 times higher than men (95% confidence interval [CI], ). Overall incidence rates were 3.6 for all whites (3.4 for men and 3.9 for women) and 7.1 for all blacks (6.9 for men and 7.4 for women). Incidence among blacks was almost 2 times higher than among whites (95% CI, ). When analyzed by registry, overall incidence rates were highest in the Detroit (Metro) (5.6; 95% CI, ; p <.01) and Utah (6.3; 95% CI, ; p <.01) regions when compared with the reference category (Figure 1). Age-adjusted annual incidence of primary DFSP dropped from 4.1 in 2000 to 3.1 in 2010, but timetrend analysis showed that there was no significant change in incidence over the decade. Incidence among women was greater than or equal to incidence among men in all years except Time-trend analysis showed incidence among men declined from 2000 to 2006 with an APC of 24.0% (95% CI, 25.9% to 22.0%) but then leveled off between 2006 and Incidence among women remained steady over the entire decade. Incidence among blacks was higher than whites for all years during the decade (Figure 2). There was no change in incidence in either racial group over the decade. Age-specific incidence of primary DFSP is presented in Figure 3. The highest age-specific annual incidence rates (7.0) were observed between the ages of 35 and 44 years when the population was analyzed as a whole (Figure 3). When analyzed by race, incidence among blacks was higher than whites in all age groups, with a peak incidence of 10.4 occurring in blacks between ages 45 and 49 years (Figure 3). When men and women were analyzed separately, peak incidence still occurred in the third and fourth decades for each sex. Among the younger population (ages, years), women had a higher incidence compared with men, 42:1S:JANUARY SUPPLEMENT 2016 S27

5 EPIDEMIOLOGY OF DFSP Figure 1. Incidence per million person-years of DFSP in the United States ( ) by region (adapted from seer. cancer.gov). Adaptations are themselves works protected by copyright. So in order to publish this adaptation, authorization must be obtained both from the owner of the copyright in the original work and from the owner of copyright in the translation or adaptation. but among people older than 75 years, men had a higher incidence compared with women. Five-year observed survival was 96.2% (95% CI, ) for all cases, whereas 5-year relative survival rate was 99.8% (95% CI, ). Ten-year observed survival was 91.1% (95% CI, ) for all cases, whereas 10-year relative survival rate was 99.1% (95% CI, ). Hazard ratios and 95% CIs for DFSP controlling for age, race, sex, and anatomic site are shown in Table 2. Increasing age was associated with increased mortality (Table 2). Men with DFSP were at 1.5 times higher risk for all-cause death than women (HR = 1.5; 95% CI, ), and blacks with DFSP were at 1.7 times higher risk for all-cause death than whites (HR = 1.7; 95% CI, ) (Table 2). Anatomic location of the upper limb, lower limb, and head were associated with Figure 2. Dermatofibrosarcoma protuberans: age-adjusted incidence per million person-years according to race and calendar year at diagnosis ( , SEER program studying 18 geographic areas). S28 DERMATOLOGIC SURGERY

6 KREICHER ET AL Figure 3. Dermatofibrosarcoma protuberans: age-specific incidence per million person-years according to gender ( , SEER program studying 18 geographic areas). higher mortality as compared with the most common location of the trunk (Table 2). Using the cause-specific death variable in SEER*Stat 8.0.4, 92 cases of DFSP were identified to have died of DFSP. The characteristics of these 92 patients whose cause of death is DFSP are shown in Table 3. About 41.3% of deaths were diagnosed between the ages of 40 and 59 years, 69.6% were white, 54.3% were male, and 34.8% had tumors on their trunk (Table 3). TABLE 2. Hazard Ratios and 95% CIs for DFSP Controlling for Age, Race, Sex, and Anatomic Site Characteristics HR 95% CI Age at diagnosis, yrs #19 Reference Reference $ Race Sex White Reference Reference Black Other Unknown Female Reference Reference Male Anatomic site Trunk Reference Reference Upper limb Lower limb Head Genitals Other TABLE 3. Characteristics of the 92 Patients Whose Cause of Death Is DFSP as Reported to the SEER Characteristics Age, years #19 1 (1.1) (16.3) (41.3) (27.2) $80 13 (14.1) Race Sex White 64 (69.6) Black 24 (26.1) Other 3 (3.2) Unknown 1 (1.1) Male 50 (54.3) Female 42 (45.7) Anatomic site Trunk 32 (34.8) Upper limb 15 (16.3) Lower limb 25 (27.2) Head 14 (15.2) Genitals 2 (2.2) Other 4 (4.3) 42:1S:JANUARY SUPPLEMENT 2016 S29

7 EPIDEMIOLOGY OF DFSP Discussion This study is the largest US population-based study of DFSP to date. The authors used data from 18 registries of the SEER database over an 11-year period. SEER-18 now accounts for 27.8% of the US population. Previous studies have used data from SEER-9, which now accounts for only 9.4% of the US population. Using the SEER program, the authors were able to gather a large number of cases of a rare disease and perform population-based analyses with statistical significance. Primary DFSP is a rare disease with an overall estimated age-adjusted incidence of 4.1 per million person-years. This is similar to the most recent reports of DFSP incidence in the United States. 7,8,12 Population-based studies in Canada and France have reported that the overall incidence of DFSP was 9.3 and 3.0, respectively. The authors have no explanation for these geographical differences in incidence. Although recent data have suggested that the incidence of primary DFSP is rising, specifically among females and whites, 7,8 the data show that incidence has remained steady over the last decade and has not changed for either sex or race. This is in contrast to the rapid increase in incidence of melanoma and most nonmelanoma skin cancers observed in the last few decades Recent studies of DFSP using SEER have shown no significant difference in incidence between men and women, 7,12 and multiple earlier studies reported higher incidence in men than women The authors showed that incidence among women is slightly higher than men, and this finding was statistically significant. Dermatofibrosarcoma protuberans incidence is highest among people in the third and fourth decades of life, younger than has previously been reported. 7,8 Overall incidence among blacks is even higher than previously reported 7,8,12 and is twice that of whites. This racial discrepancy was more pronounced in blacks aged 45 to 49 years, for which incidence reached 3.5 times that of the population as a whole. A recent study has shown people with DFSP to be at increased risk for female hormone related cancers, including breast and soft-tissue cancers, suggesting a possible hormonal association to the development of DFSP. 19 Hormones could explain the higher incidence of DFSP among women and among blacks, as studies have shown that blacks men and black women have higher estrogen levels than white men and white women, respectively. 20,21 The data support previous findings that the trunk is the most common anatomic location of this tumor, 7,12,22 followed by upper limb, lower limb, head, and genitals. This was true of both sexes and all age groups except men older than 80 years of age for which the proportion of cases at the head was greatest. The proportion of the tumors on the head gradually increased with age in men, but this pattern was not seen in women. Interestingly, the same pattern has been noted in men with Merkel cell carcinoma, but an explanation has not been fully elucidated. 23 Published data on DFSP survival are limited. Dermatofibrosarcoma protuberans remains a disease of low mortality with relative 10-year survival of 99.1%. This is comparable with previous reports of DFSP survival. 7 Advanced age at diagnosis decreased survival independently from sex, race, and anatomic site. The data indicate that male sex, black race, and localization on the head and limbs are negative predictors of survival, whereas localization on the trunk, the most common location, is a positive predictor of survival. The authors used a new cause-specific death classification implemented by SEER in 2008 to identify all deaths in the cohort attributable to DFSP. The code allows them to capture all deaths related to a specific cause but not necessarily coded as such. One previous SEER study of DFSP described only 8 deaths caused by nonmelanoma skin cancer of 2,885 DFSP cases. 7 The authors were able to capture 92 deaths attributable to DFSP of 6,817 total DFSP cases, suggesting that there are more deaths that can be attributed to death by DFSP than originally thought. Limitations of this study include lack of verification of individual diagnoses and limited follow-up on each case in the SEER program. There also may be differences in reporting and follow-up among the 18 different SEER registries. S30 DERMATOLOGIC SURGERY

8 KREICHER ET AL In conclusion, this is the largest population-based study of DFSP to date. Incidence of DFSP has remained steady over the last decade and is highest among blacks and women. Dermatofibrosarcoma protuberans is a rare cutaneous sarcoma with unknown etiology and a demographic profile that differs from other skin malignancies. Recognition of these differences may provide insight into the etiology of DFSP in hopes of preventing this disease in the future. References 1. McArthur GA, Demetri GD, van Oosterom A, Heinrich MC, et al. Molecular and clinical analysis of locally advanced dermatofibrosarcoma protuberans treated with imatinib: imatinib target exploration consortium study B2225. J Clin Oncol 2005;23: Simon MP, Pedeutour F, Sirvent N, Grosgeorge J, et al. Deregulation of the platelet-derived growth factor B-chain gene via fusion with collagen gene COL1A1 in dermatofibrosarcoma protuberans and giant-cell fibroblastoma. Nat Genet 1997;15: Sirvent N, Maire G, Pedeutour F. Genetics of dermatofibrosarcoma protuberans family of tumors: from ring chromosomes to tyrosine kinase inhibitor treatment. Genes Chromosomes Cancer 2003;37: Bowne W, Antonescu C, Leung D, Katz S, et al. Dermatofibrosarcoma protuberans: a clinicopathologic analysis of patients treated and followed at a single institution. Cancer 2000;88: Dawes K, Hanke C. Dermatofibrosarcoma protuberans treated with Mohs micrographic surgery: Cure rates and surgical margins. Dermatol Surg 1996;22: Rutgers E, Kroon B, Albus-Lutter C, Gortzak E. Dermatofibrosarcoma protuberans: treatment and prognosis. Eur J Surg Oncol 1992;18: Criscione VD, Weinstock MA. Descriptive epidemiology of dermatofibrosarcoma protuberans in the united states, 1973 to J Am Acad Dermatol 2007;56: Toro JR, Travis LB, Wu HJ, Zhu K, et al. Incidence patterns of soft tissue sarcomas, regardless of primary site, in the surveillance, epidemiology and end results program, : an analysis of 26,758 cases. Int J Cancer 2006;119: Surveillance, Epidemiology, and End Results (SEER) Program. SEER*Stat Database: Incidence: SEER 9 Regs Research data, November 2011 Sub ( ). Linked to County Attributes, Total US, Counties, National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2013, based on the November 2012 submission. Available from Accessed February 14, Percy C, Fritz A, Ries L. Conversion of Neoplasms by Topography and Morphology From the International Classification of Diseases for Oncology, Second Edition (ICD-O-2) to the International Classification of Diseases for Oncology, Third Edition (ICD-O-3). Bethesda, MD: National Cancer Institute; Surveillance, Epidemiology, and End Results (SEER) Program: SEER*Stat Available from: Accessed February 14, Rouhani P, Fletcher CDM, Devesa SS, Toro JR. Cutaneous soft tissue sarcoma incidence patterns in the U.S. Cancer 2008;113: Hacker SM, Flowers FP. Squamous cell carcinoma of the skin. will heightened awareness of risk factors slow its increase? Postgrad Med 1993;93: Marks R. The epidemiology of non-melanoma skin cancer: who, why and what can we do about it. J Dermatol 1995;22: Siegel R, Naishadham D, Jemal A. Cancer statistics, CA Cancer J Clin 2013;63: Gloster HM Jr. Dermatofibrosarcoma protuberans. J Am Acad Dermatol 1996;35: Mendenhall WM, Zlotecki RA, Scarborough MT. Dermatofibrosarcoma protuberans. Cancer 2004;101: Hussain SK, Sundquist J, Hemminki K. Incidence trends of squamous cell and rare skin cancers in the Swedish national cancer registry point to calendar year and age-dependent increases. J Invest Dermatol 2010; 130: Kurlander DE, Martires KJ, Chen Y, Barnholtz-SloanJS,etal.Riskof subsequent primary malignancies after dermatofibrosarcoma protuberans diagnosis: a national study. J Am Acad Dermatol 2013; 68: Rohrmann S, Nelson WG, Rifai N, Brown TR, et al. Serum estrogen, but not testosterone, levels differ between black and white men in a nationally representative sample of Americans. J Clin Endocrinol Metab 2007;92: Marsh EE, Shaw ND, Klingman KM, Tiamfook-Morgan TO, et al. Estrogen levels are higher across the menstrual cycle in africanamerican women compared with caucasian women. J Clin Endocrinol Metab 2011;96: Monnier D, Vidal C, Martin L, Danzon A, et al. Dermatofibrosarcoma protuberans: a population-based cancer registry descriptive study of 66 consecutive cases diagnosed between 1982 and J Eur Acad Dermatol Venereol 2006;20: Agelli M, Clegg LX. Epidemiology of primary Merkel cell carcinoma in the united states. J Am Acad Dermatol 2003;49: Address correspondence and reprint requests to: Kathryn L. Kreicher, BA, Department of Dermatology, University Hospitals Case Medical Center, Case Western Reserve University, Euclid Avenue, Lakeside 3500, Cleveland, OH 44106, or klk88@case.edu 42:1S:JANUARY SUPPLEMENT 2016 S31