Descriptive epidemiology of Kaposi sarcoma in Europe. Report from the RARECARE project

Cancer Epidemiology 38 (2014) 670–678

Contents lists available at ScienceDirect

Cancer Epidemiology The International Journal of Cancer Epidemiology, Detection, and Prevention journal homepage: www.cancerepidemiology.net

Descriptive epidemiology of Kaposi sarcoma in Europe. Report from the RARECARE project C.A. Stiller a,*, A. Trama b, D.H. Brewster c, J. Verne d, C. Bouchardy e, C. Navarro f,g,h, M.D. Chirlaque g, R. Marcos-Gragera i, O. Visser j, D. Serraino k, E. Weiderpass l,m,n,o, A.P. Dei Tos p, V. Ascoli q the RARECARE Working Group1 a

Childhood Cancer Research Group, Department of Paediatrics, University of Oxford, Oxford, UK Department of Preventive and Predictive Medicine, Fondazione IRCSS, Istituto Nazionale dei Tumori, Via Venezian 1, 20133 Milan, Italy c Scottish Cancer Registry, Information Services Division, National Services Scotland Gyle Square, 1 South Gyle Crescent, Edinburgh EH12 9EB, Scotland, UK d Public Health England Knowledge & Intelligence Team (South West), Grosvenor House, 149 Whiteladies Road, Bristol BS8 2RA, UK e Geneva Cancer Registry, IMSP – University of Geneva 55, Bd de la Cluse, CH-1205 Geneva, Switzerland f Department of Epidemiology, Murcia Regional Health Authority, Ronda de Levante 11, 30008 Murcia, Spain g CIBER Epidemiologı´a y Salud Pu´blica (CIBERESP), Spain h Department of Health and Social Sciences, Universidad de Murcia, Murcia, Spain i Epidemiology Unit and Girona Cancer Registry, Oncology Coordination Plan, Department of Health, Autonomous Government of Catalonia and Descriptive Epidemiology, Genetics and Cancer Prevention Group [Girona Biomedical Research Institute (IdIBGi)], Catalan Institute of Oncology, Carrer del Sol, 15, 17004 Girona, Spain j Comprehensive Cancer Centre the Netherlands, Registration & Research, PO Box 19.079, 3501 DB Utrecht, The Netherlands k SOC Epidemiologia e Biostatistica, Friuli Venezia Giulia Cancer Registry, IRCCS Centro di Riferimento Oncologico, Aviano, Italy l Department of Research, Cancer Registry of Norway, Oslo, Norway m Department of Community Medicine, Faculty of Health Sciences, University of Tromsø, The Arctic University of Norway, Tromsø, Norway n Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden o Samfundet Folkha¨lsan, Helsinki, Finland p Department of Oncology and Anatomic Pathology and General Hospital of Treviso, Piazza Ospedale 1, Treviso, Italy q Dipartimento di Scienze Radiologiche, Oncologiche ed Anatomopatologiche, Universita` Sapienza, Viale Regina Elena 324, 00161 Roma, Italy b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 9 April 2014 Received in revised form 6 August 2014 Accepted 26 September 2014 Available online 22 October 2014

Kaposi sarcoma (KS) is a virus-related malignancy which most frequently arises in skin, though visceral sites can also be involved. Infection with Kaposi sarcoma herpes virus (KSHV or HHV-8) is required for development of KS. Nowadays, most cases worldwide occur in persons who are immunosuppressed, usually because of HIV infection or as a result of therapy to combat rejection of a transplanted organ, but classic Kaposi sarcoma is predominantly a disease of the elderly without apparent immunosuppression. We analyzed 2667 KS incident cases diagnosed during 1995–2002 and registered by 75 population-based European cancer registries contributing to the RARECARE project. Total crude and age-standardized incidence rate was 0.3 per 100,000 per year with an estimated 1642 new cases per year in the EU27 countries. Age-standardized incidence rate was 0.8 per 100,000 in Southern Europe but below 0.3 per 100,000 in all other regions. The elevated rate in southern Europe was attributable to a combination of classic Kaposi sarcoma in some Mediterranean countries and the relatively high incidence of AIDS in several countries. Five-year relative survival for 2000–2002 by the period method was 75%. More than 10,000 persons were estimated to be alive in Europe at the beginning of 2008 with a past diagnosis of KS. The aetiological link with suppressed immunity means that many people alive following diagnosis of KS suffer comorbidity from a pre-existing condition. While KS is a rare cancer, it has a relatively good prognosis and so the number of people affected by it is quite large. Thus it provides a notable example of the importance of networking in diagnosis, therapy and research for rare cancers. ß 2014 Elsevier Ltd. All rights reserved.

Keywords: Rare cancers Kaposi sarcoma Incidence Survival Prevalence Europe Population-based study

* Corresponding author at: New College, Holywell Street, Oxford OX1 3BN, UK. E-mail address: [email protected] (C.A. Stiller). 1 See Appendix A for the list of RARECARE Working Group members. http://dx.doi.org/10.1016/j.canep.2014.09.009 1877-7821/ß 2014 Elsevier Ltd. All rights reserved.

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

1. Introduction Kaposi sarcoma (KS) is a virus-related malignancy which most frequently arises in skin, though mucosal sites, lymph nodes and viscera can also be involved [1]. There is a male predominance, and KS can develop at any age [1]. Infection with Kaposi sarcoma herpes virus (KSHV – previously known as human herpes virus type 8, HHV-8) is required for the development of KS [2]. A hallmark of KS is the large-scale variation in incidence rates that reflects the seroprevalence of KSHV [2], which also occurs on a small geographical scale [3,4]. Historically, KS occurred as two clinically and epidemiologically distinct subtypes, classic and endemic. Classic KS is predominantly a disease of the elderly of Mediterranean or Middle Eastern origin without apparent immunosuppression; never smoking, diabetes and use of oral corticosteroids are risk factors [5,6]. Geographical variations in KSHV seroprevalence and the incidence of classic KS remain largely unexplained [7]. Endemic KS occurs almost entirely in sub-Saharan Africa [8]. Nowadays, most cases of KS worldwide occur in persons who are immunosuppressed [9], usually because of HIV infection or, less commonly, as a result of therapy to combat rejection of a transplanted organ [10]. Consequently, geographical variation in incidence also reflects variation in HIV prevalence [11], and it is difficult to disentangle endemic KS and HIV-related KS in sub-Saharan Africa. KS is rare in developed countries and there is a paucity of comparative population-based information on incidence [12] and survival. This is partly because it was not visible in publications in which tumours were grouped according to the International Classification of Diseases (ICD). It was only following the AIDS epidemic that KS was assigned its own code in ICD-10 (C46). The aim of this paper is to describe incidence, prevalence and survival for KS in Europe, based on a large population-based series of data collected from 89 European cancer registries (CRs) in the framework of the project ‘Surveillance of rare cancer in Europe’ (RARECARE).

2. Materials and methods RARECARE gathered data on cancer patients diagnosed from 1978 to 2002 from 89 population-based CRs, with vital status information available up to at least 31 December 2003. In the present paper we considered data from 75 CRs in 21 countries, excluding those which did not classify cancers according to the third edition of the International Classification of Diseases for Oncology (ICD-O-3) and also those which collected data on childhood cancers only. Eleven countries were covered by national CRs (Austria, Iceland, Ireland, Malta, Norway, Slovakia, Slovenia, Sweden, Northern Ireland, Scotland and Wales), while the other ten countries (Belgium, England, France, Germany, Italy, Poland, Portugal, Spain, Switzerland, The Netherlands) were represented by regional CRs. The registries were grouped by country and the countries were grouped into regions: Northern Europe (Iceland, Norway, Sweden), United Kingdom and Ireland (England, Scotland, Wales, Northern Ireland, Republic of Ireland), Central Europe (Belgium, Austria, France, Germany, The Netherlands, Switzerland), Eastern Europe (Poland, Slovakia), and Southern Europe (Italy, Malta, Portugal, Slovenia, Spain). We present data on KS (ICD-O-3 morphology code 9140 combined with any T code). The incidence analysis considered cases incident from 1995 to 2002 and excluded 12 specialized CRs that did not cover all cancer sites. Thus, the incidence analyses were restricted to 63 CRs. Incidence rates were estimated as the number of new cases occurring in 1995–2002 divided by the total person-years in the general population in the registry areas considered, over the same

671

period. For age-standardized rates, the European population was used. The prevalence per 100,000 was estimated at the index date of 1st January 2003. The counting method [13], was applied to CR data from 1988 to 2002. Only data from 22 CRs, covering the whole 15-year period, were used for prevalence estimation. The completeness index method [14] was used to estimate complete prevalence, and involved adding the estimated surviving cases diagnosed before 1988 to those counted in 1988–2002. The expected number of new cases in the European Union (EU27) in 2008 was estimated by multiplying the crude incidence rate by the 2008 EU27 population (497,455,033) provided by EUROSTAT. The expected number of prevalent cases in 2008 was estimated by multiplying the prevalence per 100,000 by the same population. We assumed that the population covered by our CRs was representative of the population of the EU27 as a whole. Relative survival was estimated as the ratio of observed survival to the expected survival in the general population of the same age and sex. We used the Ederer II method, multiplication of the annual relative survival ratios, because this tends to be less prone to overestimation of relative survival than other methods [15]. Relative survival refers to the period 2000–2002 and was estimated by the period approach [16]. Overall 46 CRs contributed to survival estimation. We estimated differences in survival time with the period survival method [16] for four follow-up periods: 1991– 1993 (period of diagnosis Jan 1, 1987–Dec 31, 1993), 1994–1996 (period of diagnosis Jan 1, 1990–Dec 31, 1996), 1997–1999 (period of diagnosis Jan 1, 1993–Dec 31, 1999), 2000–2002 (period of diagnosis Jan 1, 1996–Dec 31, 2002); 22 CRs were included in the time trend analysis. Cases ascertained by death certificate only (DCO) were excluded from the survival analyses. 3. Results The analyses were carried out on 2667 cases that were diagnosed during 1995–2002. The numbers of registrations and main data quality indicators are presented in Table 1. Overall, 0.8% of cases were DCO. Less than 0.5% of cases were diagnosed at autopsy. Microscopic verification was present in 91% of cases. The proportion of cases diagnosed during 1995–1998 that were censored before 5 years was below 2%. The population covered by all CRs listed in Table 1 was 123,155,027, corresponding to 25% of the EU27 population. 3.1. Incidence Table 2 shows the crude incidence rates for KS overall and by sex and age group. Total age-standardized incidence was the same as the crude incidence of 0.3/100,000 per year. There were 1642 new diagnoses expected in the EU27 in 2008. Table 3 shows incidence by primary site. Skin and soft tissue are combined because some CRs routinely coded KS of the skin to soft tissue (C49). Overall, 2272 registrations (85%) had skin or soft tissue as primary site, 95 (3.6%) were in the oral cavity and pharynx, 49 (1.8%) in the digestive system, 29 (1.1%) in the penis, 18 (0.7%) in the respiratory system and 204 (7.6%) were in other and unspecified sites. The age group 0–44 years accounted for 35% (944/2272) of the registrations for all sites combined, but 72% (13/ 18) in the respiratory system and 67% (64/95) in the oral cavity and pharynx. Incidence in males was more than three times that in females. Overall incidence rose steadily with age until 35–39, declined to a local minimum at age 50–54 and rose steeply into old age (Fig. 1A). The peak in early adulthood was much more marked in males than in females. The age-incidence distribution varied between regions (Fig. 1B and C). The early peak was most

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

672

Table 1 Countries and cancer registries participating in RARECARE with mean population size covered by registration in 1995–2002 and proportion (%) of national population covered and data quality indicators of Kaposi sarcoma diagnosed in all RARECARE cancer registries, cases diagnosed 1995–2002. Countries with national coverage in bold. Mean population

Registry

Austria (national) Belgium (Flanders) France (Bas-Rhin, Calvados, Calvados digestive, Coˆte d’Or digestive, Coˆte d’Or haematol, Doubs, Haut-Rhin, He´rault, Ise`re, Loire Atlantique, Manche, Marne & Ardennes, Somme, Tarn) Germany (Saarland) Iceland Ireland Italy (Alto Adige, Biella, Ferrara, Firenze, Friuli V.G., Genova, Macerata, Modena, Napoli, Palermo, Parma, Ragusa, Reggio Emilia, Romagna, Salerno, Sassari, Trento, Umbria, Varese, Veneto) Malta Netherlands (Amsterdam, Eindhoven, North Netherlands) Norway Poland (Warsaw and Kielce) Portugal (South Portugal) Slovakia (national) Slovenia (national) Spain (Albacete, Basque Country, Castillon, Girona, Granada, Murcia, Navarra, Tarragona) Sweden Switzerland (Basel, Geneva, Grisons, St Gallen, Ticino, Valais, Zurich) UK-England (East Anglia, Northern & Yorkshire, Oxford, South West, Trent, West Midlands) UK-Northern Ireland UK-Scotland UK-Wales European CRs

% national coverage

Death certificate only (%)

Number of cases 1995–2002

Autopsy (%)

Microscopic verification (%)

Cases 1995–1998 censored before 5 years (%)

7,963,020 5,919,586 8,652,004

100 58.2 10.5–14.7

38 59 20

0.0 0.0 0.0

0.0 0.0 0.0

92.1 91.5 65.0

0.0 0.0 5.0

1,079,880 270,581 3,659,684 14,083,853

1.3 100 100 23.7–25.8

8 35 37 1183

0.0 0.0 0.0 0.6

0.0 0.0 2.7 0.0

87.5 100.0 81.1 94.7

12.5 0.0 0.0 1.9

373,866 5,370,176 4,394,802 2,799,104 4,401,902 5,381,037 1,985,998 6,456,423

100 34.1 100 7.3 43.4 100 100 12.2–16.3

23 261 105 8 113 7 8 185

0.0 0.0 0.0 0.0 0.0 14.3 0.0 0.5

0.0 1.1 0.0 0.0 0.0 0.0 0.0 0.0

87.0 88.5 96.2 100.0 100.0 85.7 100.0 89.2

0.0 1.9 1.0 0.0 0.0 0.0 0.0 1.6

8,840,065 3,334,008

100 27.1–46.8

243 98

0.0 0.0

1.2 1.0

100.0 81.6

0.8 3.1

28,534,991

57.8

172

5.2

0.0

70.3

1.7

1,667,784 5,085,648 2,900,615 123,155,027

100 100 100

6 44 14 2667

0.0 0.0 21.4 0.8

0.0 0.0 0.0 0.3

66.7 81.8 64.3 91.5

0.0 0.0 0.0 1.6

Table 2 Kaposi sarcoma in Europe. Observed cases with overall, sex-specific and age-specific incidence rates/100,000 with standard errors (SE). RARECARE cancer registries (1995– 2002). Europe and European region

Rate SE observed cases 1995–2002

Europe 2667 Northern Europe 383 Central Europe 483 Eastern Europe 15 Southern Europe 1513 UK and Ireland 273

0.33 0.25 0.24 0.02 0.80 0.09

Sex

Age

Male

Female

0–44 years

45–64 years

65+ years

Rate SE observed cases 1995–2002

Rate SE observed cases 1995–2002

Rate SE observed cases 1995–2002

Rate SE observed cases 1995–2002

Rate SE observed cases 1995–2002

0.01 2080 0.01 268 0.01 424 0.01 9 0.02 1137 0.01 242

0.49 0.41 0.43 0.04 1.36 0.16

0.01 587 0.03 115 0.02 59 0.01 6 0.04 376 0.01 31

pronounced in Southern Europe and absent in Eastern Europe. The rise in old age was also most marked in Southern Europe, moderate in Northern Europe and very slight in the other regions. Age-standardized rates were 0.8/100,000 in Southern Europe and below 0.3 in all other European regions. There was greater variation between countries, both in total incidence and in the pattern by age group (Fig. 2). The highest overall ASRs, above 1/ 100,000, were observed in Iceland and Portugal, while several countries had rates below 0.1/100,000. In the age-group 0–44 years, Portugal had the highest rate, 1.51/100,000, based on 77 cases; Italy (N = 310 cases), the Netherlands (N = 160), Spain

0.10 0.11 0.05 0.02 0.31 0.02

0.00 0.01 0.01 0.01 0.02 0.00

944 62 246 0 473 163

0.20 0.10 0.22 0.00 0.57 0.09

0.01 603 0.01 63 0.01 142 – 6 0.03 338 0.01 54

0.32 0.24 0.32 0.04 0.93 0.08

0.01 1120 0.03 258 0.03 95 0.02 9 0.05 702 0.01 56

0.89 1.43 0.34 0.11 2.68 0.12

0.03 0.09 0.03 0.04 0.1 0.02

(N = 81) and Switzerland (N = 43) all had rates in the range 0.48– 0.60/100,000, and all other countries had rates below 0.2/100,000. Incidence at age 45–64 was between 0.6 and 1.1/100,000 in Italy (N = 255), the Netherlands (N = 80), Portugal (N = 22), Spain (N = 52) and Switzerland (N = 28), and also in Iceland (though based on only 3 cases). Incidence in this age group was below 0.3/ 100,000 in all other countries. Iceland had exceptionally high incidence at age 65+, above 12/100,000, based on 31 cases. Elsewhere, the highest incidence in this age group was observed in Italy (N = 620) and Malta (N = 13), both above 3/100,000. Several other countries had incidence around 1/100,000 at age 65+.

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

673

Table 3 Kaposi sarcoma in Europe. Incidence (period of diagnosis 1995–2002) by age group and primary site. All ages

Oral cavity and pharynx Digestive system Respiratory system Skin Penis Other and unspecified sitesa

00–44 years

45–64 years

65+ years

Rate

SE

Observed cases

%

Rate

SE

Observed cases

%

Rate

SE

Observed cases

%

Rate

SE

Observed cases

%

0.012 0.006 0.002 0.284 0.004 0.026

0.001 0.001 0.001 0.006 0.001 0.002

95 49 18 2272 29 204

4 2 1 85 1 8

0.013 0.005 0.003 0.151 0.002 0.021

0.002 0.001 0.001 0.006 0.001 0.002

64 24 13 732 9 102

7 3 1 78 1 11

0.012 0.006 0.003 0.272 0.004 0.021

0.003 0.002 0.001 0.012 0.001 0.003

23 12 5 515 8 40

4 2 1 85 1 7

0.006 0.010 NE 0.814 0.010 0.049

0.002 0.003 NE 0.025 0.003 0.006

8 13 0 1025 12 62

1 1 NE 92 1 6

NE: not estimable. a Female genital system, prostate, other and ill-defined sites.

3.2. Prevalence At the index date of 1st January 2003, the observed prevalence proportions per 100,000 were 0.3 (SE < 0.1) at 2 years, 0.8 (SE < 0.1) at 5 years and 1.5 (SE < 0.1) at 15 years and the estimated complete prevalence in Europe was 2.1 (SE 0.1). There were estimated to be 10,516 persons alive in the EU27 at the beginning of 2008 with a past diagnosis of KS. Of these, 1700 (14%) and 3795 (38%) respectively were diagnosed within 2 and 5 years before the prevalence date. The 24% difference between these proportions represents cases in the third to fifth years following diagnosis, presumably undergoing clinical follow-up. The remaining 6721 (62%) represent long-term survivors, among whom 2896 (29% of the total) had survived more than 15 years after diagnosis. 3.3. Survival Survival for the period 2000–2002 was estimated from 1800 cases. Five-year relative survival was 75% for all primary sites combined and 76% for patients with skin as primary site (Table 4). Among the other primary sites, survival ranged from 25% for digestive system to 89% for penis, but these estimates are based on a limited number of cases (<30) (Table 4). Five-year relative survival varied little with age, being 74% for ages 0–44 and 45–64 and 76% for age 65+. Fig. 3 shows 5-year relative survival by European regions. Five-year relative survival from KS in regions with more than 30 cases for analysis was 60% in Northern Europe, 66% in the UK and Ireland, 73% in Central Europe and 82% in Southern Europe. Survival in Eastern Europe was 55% but this was based on <30 cases. Table 5 shows 5-year relative survival by age group for the periods 1991–1993, 1994–1996, 1997–1999 and 2000–2002. In 1991–1993 and 1994–1996 the age groups 0–44 and 45–64 years had much lower survival than the 65+ age group. Survival increased dramatically between 1994–1996 and 1997–1999 for the two younger age groups, but there was no sign of a trend in the 65+ group. In 1997–1999 and 2000–2002, relative survival was very similar across all three age groups. 4. Discussion Our results describe, for the first time, the epidemiology of KS throughout much of Europe. The data are from population-based CRs and have been subjected to uniform, stringent validation procedures. The information in this paper should therefore be accurate, assuming that there have been no major changes in incidence or survival since the study period. KS had an incidence of 0.3/100,000, with about 1650 new cases per year in Europe. Incidence was highest above the age of 65 years, but in southern Europe a clear peak in the age-group 35–39 was

also observed, and the highest incidence in central Europe and the UK and Ireland was between 30 and 49 years of age. Incidence in males was more than three times that in females. More than 10,000 people are estimated to be alive in Europe with a past diagnosis of KS. Incidence in Europe was markedly lower than in the United States among males (0.5/100,000 vs. 3.3/100,000), whereas incidence in females was 0.1/100,000 in Europe and the United States [17]. Epidemiological subtype of KS is not generally recorded by CRs and was not included in the data analyzed here. Primary site is a poor discriminator. Oral cavity and visceral organs are typical sites for AIDS-related KS but accounted for a small percentage of all cases. The skin, which does not discriminate between classic and AIDS-related KS, was primary site in the majority of cases. Age at diagnosis is a much better discriminator. In an Italian study, between 1985 and 1998, 92% of cases of AIDS-related KS occurred before the age of 55, while classic KS accounted for 97% of all KS in people aged 65 years and above [18]. Our data showed that sites typical for AIDS-related KS (oral cavity and viscera) were more common in the younger age group than in the older, supporting the idea that AIDS-related KS is typical of younger age group. There was marked geographical variation in the incidence of KS between and within European regions. Incidence in Southern Europe was more than twice that in any other European region and there was also a peak in the age-group 30–39 (supporting the role of HIV-related KS in these countries). Italian registries provided the majority of cases within this region and the elevated national rate reflected the substantially higher incidence of classic KS in southern Italy and parts of Lombardy [3,18] and of AIDS-associated KS in northern and central Italy [19]. Spain and Portugal, the other large contributors of data from Southern Europe, do not have a high incidence of classic KS. Until the mid 1990s Spain had the highest incidence of AIDS among European countries but infection rates have been falling since then [20] and this should lead to a corresponding fall in the incidence of KS. Portugal had the highest national incidence of AIDS in Europe in the early 2000s [21] and this is reflected in its exceptionally high rates of KS in the age groups 0–44 and 45–64 years. Of the other contributors in Southern Europe, Malta had a high incidence for classic KS [22], and it had one of the highest incidence rates in the age group 65+. By contrast, the rate in Slovenia was among the lowest observed, suggesting that the epidemiology of KS there more closely resembles that in Eastern Europe. Both countries have low incidence of AIDS [20]. Among countries of Southern Europe not included in this study, the incidence of classic KS in Greece was 0.4/ 100,000 [23]. Within Northern Europe, Iceland has an exceptionally high incidence of classic KS [12,24], and in the present study it had the highest incidence among the 65+ age group. These were

674

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

Fig. 1. Age-sex-specific incidence rates of Kaposi sarcoma in Europe (A) and by European region for male (B) and female (C). Period of diagnosis 1995–2002.

predominantly skin tumours (94%) and 82% were in people aged 70 years and above, so the high incidence might reflect efficient surveillance of the elderly population in Iceland. Similarly high incidence of classic KS has been reported in the Faroe Islands, not included in the present study [12,24]. In Central Europe and UK/Ireland, incidence of KS was highest at age 30–49, consistent with most cases being AIDS-related. The highest national incidence estimates in Central Europe were in Switzerland and the Netherlands, with considerably lower rates in

Belgium, France, Austria and Germany. Similar, relatively low rates were observed in the UK and Ireland. Among these eight countries, Switzerland and France had the highest incidence of AIDS in 2000 (30.1 and 28.7 per million, respectively) [21]. Incidence of AIDS was considerably lower in the remaining six countries, ranging from 15.6 per million in the Netherlands to 3.4 per million in Ireland [21]. Within these regions, however, this study included national cancer data only for Austria and Ireland, with the remaining countries being incompletely represented by regional

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

675

Fig. 2. Standardized incidence rates of Kaposi sarcoma in Europe by country and age groups: (A) all ages; (B) age 00–44; (C) age 45–64; (D) age 65+. Period of diagnosis 1995– 2002. Error bars are 95% CI.

registries. In particular, the proportions of the national populations of France and Germany covered in this study are very small and the data may not be typical of the countries as a whole. The years covered by the incidence data in this study roughly correspond to the early-highly active antiretroviral therapy (HAART) period [25]. In Switzerland, the incidence of KS among people with AIDS fell by 86% from 1375/100,000 pre-HAART (1985–1996) to 67 in the late-HAART period (2002–2006) [25], reflecting the reduced risk of KS in patients receiving HAART. In Italy, the incidence of KS among people with AIDS fell by 88%, from 2131/100,000 to 250, between 1986–1996 and 1997–2004 [26].

Table 4 Kaposi sarcoma in Europe. Estimated 5-year relative survival rates and standard errors (SE) by primary site. Period survival analysis 2000–2002. The number of cases is estimated on 7 years of incidence. Primary site

All sites Oral cavity and pharynx Digestive system Respiratory system Skin Penis Other and unspecified sites

5-year relative survival No cases (1996–2002)

%

SE

1795 43 27 11 1553 25 136

74.9 76.5 25.3 50.4 75.5 88.7 73.2

2.2 11.7 17.1 35.6 2.3 14.8 7.6

This study included only two countries of Eastern Europe, namely Poland and Slovakia, both with very low incidence. The rarity of KS at younger ages in Poland and Slovakia is consistent with the low incidence of AIDS in both countries [21]. Incidence of classic KS correlates with seroprevalence of KSHV [2]; high levels of both classic KS and HHV-8 are observed in several isolated populations. Environmental risk factors may also be important [27,28]. Associations of classic KS with exposure to specific soil types have not been consistent between studies [29,30]. We estimated that at the beginning of 2008 there were about 10,500 people alive in the EU27 with a past diagnosis of KS. In a previous study using the same methodology, the estimated complete prevalence of KS for Italy was 11.91/100,000 [31]. Multiplying this by the Italian population of about 60 million gives a number of 7200 prevalent cases. This implies that almost 70% of prevalent cases of KS in the EU27 population are in Italy, consistent with the fact that Italy is by far the largest European country with historically high incidence of KS, and one of the main countries contributing to KS incidence in Europe. We observed contrasting patterns of survival by age group between 1993 and 2002. Among people diagnosed under the age of 65, in whom AIDS-related KS would have been the predominant subtype, 5-year relative survival increased markedly from 17–19% (age 0–44) and 25–39% (age 45–64) in 1993–1996 to around 70% in 1999–2002. Among those aged 65 and over, most of whom would

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

676

Fig. 3. Five-year relative survival of Kaposi sarcoma by EU Regions. Period survival analysis 2000–2002. Error bars are 95% CI. Table 5 Kaposi sarcoma in Europe. Five-year relative survival from 1991 to 2002. N (1987–2002)

Kaposi sarcoma 00–44 years 45–64 years 65+ years

1147 572 852

% survival (95% CI) 1991–1993

1994–1996

1997–1999

2000–2002

17 (13–22) 25 (18–33) 74 (58–85)

19 (15–24) 39 (30–48) 71 (57–82)

67 (59–74) 70 (59–79) 73 (58–83)

71 (62–78) 72 (61–81) 66 (53–76)

have had classic KS, 5-year relative survival was in the range 65– 75% throughout the study period. KS is an opportunistic tumour and restoration of immunity is the best way to treat epidemic and iatrogenic KS. The AIDS-related subtype which occurs predominantly in younger adults is aggressive. In the pre-HAART era, it had a correspondingly poor prognosis. Introduction of HAART saw not only a reduction in the risk of AIDS-related KS but also improved survival of patients with the disease [32]. Variations in timely diagnosis and in access to appropriate treatment, particularly for AIDS-related and iatrogenic KS, might have contributed to the inequality of survival between European regions but these factors cannot be studied with the data available here. Classic KS has a more indolent behaviour and is generally chemosensitive. However, limited information is available about the duration of the response, who needs treatment and when treatment should be started. A review concluded that evidence does not support recommending any particular therapeutic strategy [33]. Only 2 randomized trials on classic KS have been carried out in the last three decades [33]. Classic KS is relatively rare and only small numbers of patients are treated at each centre. The best way to improve treatment would be to establish collaborative networks for the planning and running of international clinical trials in the near future, making it possible to identify the most effective treatments. Survival of patients with KS in 2000–2002 varied between European regions, being higher in Southern and Central Europe than elsewhere. Even though survival did not vary with age, the geographical variation in occurrence of KS subtypes could still have contributed to the difference in outcome between regions. Higher survival in Southern Europe may partially reflect predominantly less aggressive disease in patients with classic KS [11] and greater clinical experience as a consequence of the higher incidence of

both major subtypes of KS. The survival estimate for Eastern Europe was based on a small number of cases and had a correspondingly wide confidence interval. This study could provide only limited information on the prognostic importance of primary site in KS because of the rarity of sites other than skin and soft tissue. However, the poor survival of visceral KS can be explained by the fact that these sites are more often affected in HIV-related KS which has a worse prognosis. The annual incidence of KS in Europe is 0.3/100,000 with 1600 new cases occurring every year. The high survival rate means that the number of people affected by it is quite large. Thus it provides a notable example of the importance of networking in diagnosis, therapy and research for rare cancers. Funding This research was supported by the European Commission through the Executive Agency for Health and Consumers (Grant No. 2006113), and the Programma Italia-USA Malattie Rare (Grant No. 526D/42). During the period of this research, the Childhood Cancer Research Group received funding from the Department of Health, the National Cancer Intelligence Network, the Scottish Government and Children with Cancer UK. The views expressed here are those of the authors and not necessarily those of the Department of Health, the National Cancer Intelligence Network, the Scottish Government, Children with Cancer UK or The Cancer Registry of Norway. Conflict of interest statement The authors declare no conflicts of interest. The founding sources had no role in study design, data collection, data analysis,

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

data interpretation, in writing this report, or in the decision to submit for publication. Authorship contribution C.A. Stiller: designed the study and wrote the manuscript. A. Trama: made the analyses and contributed to write the manuscript. D.H. Brewster, J. Verne, C. Bouchardy, C. Navarro, M.D. Chirlaque, R. Marcos Gragera, O. Visser, D. Serraino, E. Weiderpass, A.P. Dei Tos: contributed to write the paper, to inform about relevant bibliographic references and experiences, to revise the manuscript. V. Ascoli: contributed to design the study and to write and revise the paper.

Appendix A. RARECARE Working Group The RARECARE Working Group consists of: Austria: N Zielonk (Austrian National Cancer Registry); Belgium: E Van Eycken (Belgian Cancer Registry); H Sundseth (European Cancer Patient Coalition); France: G Hedelin (Bas-Rhin Cancer Registry); AM Bouvier (Coˆte d’Or Digestive Cancer Registry); AS Woronoff (Doubs Cancer Registry); A Buemi (HautRhin Cancer Registry); B Tretarre (He´rault Cancer Registry); M Colonna (Ise`re Cancer Registry); S Bara (Manche Cancer Registry); O Ganry (Somme Cancer Registry); P Grosclaude (Tarn Cancer Registry); S Baconnier (Connective tissue cancer network – CONTICANET); Germany: B Holleczek (Saarland Cancer Registry); J Geissler (CML Advocates Network); M Wartenberg (Global GIST Network); Iceland: L Tryggvadottir (Icelandic Cancer Registry); Ireland: S Deady (National Cancer Registry of Ireland); Italy: F Bellu` (Alto Adige Cancer Registry); S Ferretti (Ferrara Cancer Registry); D Serraino (Friuli Venezia Giulia Cancer Registry); M Vercelli (Liguria Cancer Registry c/o IST/UNIGE, Genoa); S Vitarelli (Macerata Province Cancer Registry); M Federico (Modena Cancer Registry); M Fusco (Napoli Cancer Registry); M Michiara (Parma Cancer Registry); A Giacomin (Piedmont Cancer Registry, Province of Biella); R Tumino (Cancer Registry and Histopathology Unit, ‘‘M.P. Arezzo’’ Civic Hospital, Ragusa); L Mangone (Servizio Interaziendale di Epidemiologia, Azienda Unita` Sanitaria Locale and IRCCS, Arcispedale Santa Maria Nuova, Reggio Emilia); F Falcini (Romagna Cancer Registry); G Senatore (Salerno Cancer Registry), M Budroni (Sassari Cancer Registry); S Piffer (Trento Cancer Registry); E Crocetti (Tuscan Cancer Registry); F La Rosa (Umbria Cancer Registry); G Tagliabue (Registry of the Varese province, Fondazione IRCCS Istituto Nazionale dei Tumori, Milano); P Zambon (Veneto Cancer Registry); G Gatta, A Gronchi, L Licitra, M Ruzza, S Sowe (Fondazione IRCCS Istituto Nazionale dei Tumori); R Capocaccia, R De Angelis, S Mallone, A Tavilla (Centro Nazionale di Epidemiologia, Istituto Superiore di Sanita`); AP Dei Tos (Local Health Unit No. 9, Region of Veneto); Malta: K England (Malta National Cancer Registry); Norway: G Ursin (Cancer Registry of Norway); Poland: J Rachtan (Cracow Cancer Registry); S Gozdz (Kielce Cancer Registry); M Zwierko (Warsaw Cancer Registry); M Bielska-Lasota (National Institute of Public Health – National Institute of Hygiene, Warsaw); J Slowinski (Department of Neurosurgery in Sosnowiec, Medical University of Silesia); Portugal: A Miranda (Southern Portugal Cancer Registry); Slovakia: Ch. Safaei Diba (National Cancer Registry of Slovakia); Slovenia: M Primic-Zakelj (Cancer Registry of Slovenia); Spain: A Mateos (Albacete Cancer Registry); J Bidaurrazaga (Basque Country Cancer Registry); A Torrella-Ramos (Castillon Cancer Registry); R Marcos-Gragera (Epidemiology Unit and Girona Cancer Registry, Oncology Coordination Plan, Department of Health, Autonomous

677

Government of Catalonia and Descriptive Epidemiology, Genetics and Cancer Prevention Group); E Ardanaz (Navarra Cancer Registry); J Galceran (Tarragona Cancer Registry); C MartinezGarcia, MJ Sanchez Perez, JM Melchor (Escuela Andaluza de Salud Pu´blica); Sweden: J Adolfsson (Stockholm-Gotland Cancer Registry); M Lambe (Uppsala Regional Cancer Registry); TR Mo¨ller (Lund University Hospital); U Ringborg (Karolinska Institute); Switzerland: G Jundt (Basel Cancer Registry); M Usel (Geneva Cancer Registry); SM Ess (St. Gallen Cancer Registry); A Spitale (Ticino Cancer Registry); I Konzelmann (Valais Cancer Registry); JM Lutz (National Institute for Cancer Epidemiology and Registration); The Netherlands: JWW Coebergh (Eindhoven Cancer Registry), O Visser, R Otter, S Siesling, JM van der Zwan (Comprehensive Cancer Centre the Netherlands), H Schouten (University of Maastricht); UK-England: DC Greenberg (Eastern Cancer Registration and Information Centre); J Wilkinson (Northern and Yorkshire Cancer Registry); M Roche (Oxford Cancer Intelligence Unit); J Verne (South West Public Health Observatory); D Meechan (Trent Cancer Registry); G Lawrence (West-Midlands Cancer Intelligence Unit); J Mackay (University College of London); UK-Northern Ireland: A Gavin (Northern Ireland Cancer Registry); UK-Scotland: DH Brewster (Scottish Cancer Registry); I Kunkler (University of Edinburgh); UK-Wales: C White (Welsh Cancer Intelligence & Surveillance Unit).

References [1] Radu O, Pantanowitz L. Kaposi sarcoma. Arch Pathol Lab Med 2013;137(2): 289–94. [2] IARC monographs on the evaluation of carcinogenic risks to humans. Biological agents, vol. 100B. France: IARC Lyon, 2012: 169–95. [3] Ascoli V, Bellil S, Benedetti M, Trinca S, Ricci P, Comba P. High incidence of classic Kaposi’s sarcoma in Mantua, Po Valley, Northern Italy (1989–1998). Br J Cancer 2001;85(3):379–82. [4] Tanzi E, Zappa A, Caramaschi F, Amendola A, Lasagna D, Gatti L, et al. Human herpesvirus type 8 infection in an area of Northern Italy with high incidence of classical Kaposi’s sarcoma. J Med Virol 2005;76(4):571–5. [5] Goedert JJ, Vitale F, Lauria C, Serraino D, Tamburini M, Montella M, et al. Risk factors for classical Kaposi’s sarcoma. J Natl Cancer Inst 2002;94(22):1712–8. [6] Anderson LA, Lauria C, Romano N, Brown EE, Whitby D, Graubard BI, et al. Risk factors for classical Kaposi sarcoma in a population-based case–control study in Sicily. Cancer Epidemiol Biomarkers Prev 2008;17(12):3435–43. [7] Rohner E, Wyss N, Trelle S, Mbulaiteye SM, Egger M, Novak U, et al. HHV-8 seroprevalence: a global view. Syst Rev 2014;3:11. [8] Cook-Mozaffari P, Newton R, Beral V, Burkitt DP. The geographical distribution of Kaposi’s sarcoma and of lymphomas in Africa before the AIDS epidemic. Br J Cancer 1998;78(11):1521–8. [9] Parkin DM. The global health burden of infection-associated cancers in the year 2002. Int J Cancer 2006;118(12):3030–44. [10] Bouvard V, Baan R, Straif K, Grosse Y, Secretan B, El Ghissassi F, et al. A review of human carcinogens – Part B: Biological agents. Lancet Oncol 2009;10(4): 321–2. [11] Ebrahim SH, Peterman TA, Zaidi AA, Hamers FF. Geography of AIDS-associated Kaposi’s sarcoma in Europe. AIDS 1997;11(14):1739–45. [12] Dal Maso L, Franceschi S, Lo Re A, La Vecchia C. Comparison of the distribution of non-AIDS Kaposi’s sarcoma and non-Hodgkin’s lymphoma in Europe. Br J Cancer 1999;79(1):161–3. [13] Capocaccia R, Colonna M, Corazziari I, De Angelis R, Francisci S, Micheli A, et al. Measuring cancer prevalence in Europe: the EUROPREVAL project. Ann Oncol 2002;13(6):831–9. [14] Capocaccia R, De Angelis R. Estimating the completeness of prevalence based on cancer registry data. Stat Med 1997;16(4):425–40. [15] Hakulinen T, Seppa¨ K, Lambert PC. Choosing the relative survival method for cancer survival estimation. Eur J Cancer 2011;47(14):2202–10. [16] Brenner H, Gefeller O. An alternative approach to monitoring cancer patient survival. Cancer 1996;78(9):2004–10. [17] Rouhani P, Fletcher CD, Devesa SS, Toro JR. Cutaneous soft tissue sarcoma incidence patterns in the U.S.: an analysis of 12,114 cases. Cancer 2008;113(3):616–27. [18] Dal Maso L, Polesel J, Ascoli V, Zambon P, Budroni M, Ferretti S, et al. Classic Kaposi’s sarcoma in Italy, 1985–1998. Br J Cancer 2005;92(1):188–93. [19] Serraino D, Angeletti C, Carrieri MP, Longo B, Piche M, Piselli P, et al. Kaposi’s sarcoma in transplant and HIV-infected patients: an epidemiologic study in Italy and France. Transplantation 2005;80(12):1699–704. [20] Castilla J, Sobrino P, Lorenzo JM, Moreno C, Izquierdo A, Lezaun ME, et al. Present situation and future perspectives of the epidemic of HIV and AIDS in Spain. An Sist Sanit Navar 2006;29(1):13–25.

678

C.A. Stiller et al. / Cancer Epidemiology 38 (2014) 670–678

[21] European Centre for Disease Prevention and Control/WHO Regional Office for Europe. HIV/AIDS surveillance in Europe 2008. Stockholm: European Centre for Disease Prevention and Control, 2009. [22] Vitale F, Briffa DV, Whitby D, Maida I, Grochowska A, Levin A, et al. Kaposi’s sarcoma herpes virus and Kaposi’s sarcoma in the elderly populations of 3 Mediterranean islands. Int J Cancer 2001;91(4):588–91. [23] Touloumi G, Kaklamanis L, Potouridou I, Katsika-Hatziolou E, Stratigos J, Mueller N, et al. The epidemiologic profile of Kaposi’s sarcoma in Greece prior to and during the AIDS era. Int J Cancer 1997;70(5):538–41. [24] Hjalgrim H, Tulinius H, Dalberg J, Hardarson S, Frisch M, Melbye M. High incidence of classical Kaposi’s sarcoma in Iceland and the Faroe Islands. Br J Cancer 1998;77(7):1190–3. [25] Franceschi S, Lise M, Clifford GM, Rickenbach M, Levi F, Maspoli M, et al. Changing patterns of cancer incidence in the early- and late-HAART periods: the Swiss HIV Cohort Study. Br J Cancer 2010;103(3):416–22. [26] Polesel J, Franceschi S, Suligoi B, Crocetti E, Falcini F, Guzzinati S, et al. Cancer incidence in people with AIDS in Italy. Int J Cancer 2010;127(6): 1437–45. [27] Ascoli V, Facchinelli L, Valerio L, Zucchetto A, Dal Maso L, Coluzzi M. Distribution of mosquito species in areas with high and low incidence of classic

[28]

[29]

[30]

[31]

[32]

[33]

Kaposi’s sarcoma and seroprevalence for HHV-8. Med Vet Entomol 2006;20(2):198–208. Ascoli V, Senis G, Zucchetto A, Valerio L, Facchinelli L, Budroni M, et al. Distribution of ‘promoter’ sandflies associated with incidence of classic Kaposi’s sarcoma. Med Vet Entomol 2009;23(3):217–25. Pelser C, Dazzi C, Graubard BI, Lauria C, Vitale F, Goedert JJ. Risk of classic Kaposi sarcoma with residential exposure to volcanic and related soils in Sicily. Ann Epidemiol 2009;19(8):597–601. Goedert JJ, Calamusa G, Dazzi C, Perna A, Pelser C, Anderson LA, et al. Risk of classic Kaposi sarcoma with exposures to plants and soils in Sicily. Infect Agent Cancer 2010;5(1):23. Trama A, Mallone S, Ferretti S, Meduri F, Capocaccia R, Gatta G, RITA Working Group. The burden of rare cancers in Italy: the surveillance of rare cancers in Italy (RITA) project. Tumori 2012;98(5):550–8. Biggar RJ, Engels EA, Ly S, Kahn A, Schymura MJ, Sackoff J, et al. Survival after cancer diagnosis in persons with AIDS. J Acquir Immune Defic Syndr 2005;39(3):293–9. Re´gnier-Rosencher E, Guillot B, Dupin N. Treatments for classic Kaposi sarcoma: a systematic review of the literature. J Am Acad Dermatol 2013;68(2): 313–31.