- Email: [email protected]
Differences in incidence and survival of oral cavity and pharyngeal cancers between Germany and the United States depend on the HPV-association of the cancer site
Oral Oncology 76 (2018) 8–15
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
Differences in incidence and survival of oral cavity and pharyngeal cancers between Germany and the United States depend on the HPV-association of the cancer site
T
⁎
L. Jansena, , N. Buttmann-Schweigerb, S. Listlc,d, M. Ressinge,f, B. Holleczekg, A. Katalinich, S. Luttmanni, K. Kraywinkelb, H. Brennera,j,k, for the GEKID Cancer Survival Working Group1 a
Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany German Centre for Cancer Registry Data (ZfKD), Department of Epidemiology and Health Monitoring, Robert Koch-Institute, Berlin, Germany Department of Conservative Dentistry, University Hospital Heidelberg, Heidelberg, Germany d Radboud University Medical Center, Radboud Institute for Health Sciences, Department of Dentistry – Quality and Safety of Oral Health Care, Nijmegen, The Netherlands e Cancer Registry Rhineland-Palatinate, Mainz, Germany f Institute of Medical Biostatistics, Epidemiology and Informatics, University Medical Center, Johannes Gutenberg University Mainz, Mainz, Germany g Saarland Cancer Registry, Saarbrücken, Germany h Cancer Registry of Schleswig-Holstein, Lübeck, Germany i Cancer Registry of Bremen, Leibniz-Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany j Division of Preventive Oncology, German Cancer Research Center (DKFZ), and National Center for Tumor Diseases (NCT), Heidelberg, Germany k German Cancer Consortium (DKTK), German Cancer Research Center, Heidelberg, Germany b c
A R T I C L E I N F O
A B S T R A C T
Keywords: Oral cancer Pharynx cancer HPV Survival Incidence Trend Germany United States
Introduction: The epidemiology of squamous cell oral cavity and pharyngeal cancers (OCPC) has changed rapidly during the last years, possibly due to an increase of human papilloma virus (HPV) positive tumors and successes in tobacco prevention. Here, we compare incidence and survival of OCPC by HPV-relation of the site in Germany and the United States (US). Materials and methods: Age-standardized and age-specific incidence and 5-year relative survival was estimated using data from population-based cancer registries in Germany and the US Surveillance Epidemiology and End Results (SEER) 13 database. Incidence was estimated for each year between 1999 and 2013. Relative survival for 2002–2005, 2006–2009, and 2010–2013 was estimated using period analysis. Results: The datasets included 52,787 and 48,861 cases with OCPC diagnosis between 1997 and 2013 in Germany and the US. Incidence was much higher in Germany compared to the US for HPV-unrelated OCPC and more recently also for HPV-related OCPC in women. Five-year relative survival differences between Germany and the US were small for HPV-unrelated OCPC. For HPV-related OCPC, men had higher survival in the US (62.1%) than in Germany (45.4%) in 2010–2013. These differences increased over time and were largest in younger patients and stage IV disease without metastasis. In contrast, women had comparable survival for HPVrelated OCPC in both countries. Conclusions: Strong survival differences between Germany and the US were observed for HPV-related OCPC in men, which might be explained by differences in HPV-attributable proportions. Close monitoring of the epidemiology of OCPC in each country is needed.
Introduction Oral cavity and pharyngeal cancers (OCPC) include cancers of the lip, oral cavity, and pharynx. Together, this group of cancers accounts
for over 500,000 new cancer cases per year worldwide [1]. In 2007, the WHO concluded that human papillomavirus (HPV) type 16 is a cause of some subtypes of OCPC [2]. Many epidemiological and molecular studies have meanwhile provided evidence for this causal link, particularly
Abbreviations: OCPC, squamous cell oral cavity and pharyngeal cancers; HPV, human papillomavirus; US, United States; SEER, Surveillance, Epidemiology, and End Results; DCO, death certificate only; AAPC, average annual percentage changes; 95% CI, 95% confidence interval ⁎ Corresponding author at: Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 581, Heidelberg 69120, Germany. E-mail address: [email protected] (L. Jansen). 1 Members of the GEKID Cancer Survival Working Group are listed in the acknowledgments. https://doi.org/10.1016/j.oraloncology.2017.11.015 Received 21 July 2017; Received in revised form 9 November 2017; Accepted 14 November 2017 1368-8375/ © 2017 Elsevier Ltd. All rights reserved.
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
We selected and classified OCPC cancers based on the anatomic site and etiologic relationship with HPV following two publications [8,19]. The classification was based on the International Classification of Disease for Oncology version-3 (ICD-0-3) [20]. Cancer with morphology codes 8050–8076, 8078, 8083, 8084 and 8094 (squamous cell carcinoma) with malignant behavior (behavior code 3) of the oral cavity (topography codes C01.9-C06.9) and pharynx (codes C09.0-C10.9, C12.9-C14.8) (excluding lip, salivary glands, and nasopharynx) were included and further classified in three groups: HPV-related OCPC, HPV-unrelated OCPC, and HPV-unrelated oral tongue (C02.0-C02.3, C02.8-C02.9) (Supplementary Material). Data from the Surveillance, Epidemiology, and End Results (SEER) 13 database was used to estimate incidence and survival for the US [21]. SEER-13 covers about 41 million inhabitants (13.4% of the total US population). The same selection and classification criteria as for the German dataset were applied for the US. In sensitivity analyses, analyses were conducted after restricting the US data to Whites.
for squamous cell cancers of the oropharynx, the tonsils and parts of the tongue [3]. Consequently, a paradigm shift in the understanding of OCPC has occurred. HPV-positive and HPV-negative OCPCs have different risk factor profiles [4]. Tobacco use and alcohol consumption are considered the primary risk factors for HPV-negative OCPCs. In contrast, HPV-positive OCPCs are associated with several measures of sexual behavior and marijuana use, which have not been found to be associated with HPVnegative OCPCs. It has been estimated that the combined effects of tobacco use and alcohol consumption are responsible for 70–80% of OCPCs in the United States (US) [5], while HPV was reported to be associated with 46% of oropharyngeal carcinomas [6]. However, the attributable fractions vary strongly by geographical location and calendar time due to changes in the prevalence of HPV-infection, tobacco use and alcohol consumption. In many high-income countries, the incidence of OCPC at HPV-unrelated sites has declined, presumably because of reduced tobacco and alcohol use. In contrast, incidence for HPV-related oropharyngeal cancers has risen strongly, specifically among males [7]. While in earlier studies increases in incidence for HPV-related sites were mainly observed among ages 40–59 years [8], they have recently also been observed in patients aged 65+ years with similar magnitude [9], highlighting the rapidly changing epidemiology of OCPC. Changes in survival patterns after OCPC have also been reported in various countries. In Europe, 5-year relative survival increased between 1999–2001 and 2005–2007 from 42.3 to 47.4 for tongue and lingual tonsil and from 36.7 to 40.1 for oropharynx and tonsil cancer. For oral cavity cancer, the improvement in survival was less pronounced (45.6–48.1) [10]. In the US, survival for oropharyngeal cancer rose considerably in the last years, while it was stable over time for HPVunrelated OCPC sites [8]. These differences have been explained by the increasing proportion of HPV-positive OCPCs together with the more favorable prognosis of HPV-positive OCPC [11]. Thus, the epidemiology of OCPC is changing strongly over time and will lead to modifications in OCPC management [12]. The magnitude of these changes will depend on changes in the prevalence of the main risk factors tobacco use and HPV. Here, we provide a comprehensive comparison of incidence and survival of OCPC at HPV-related and non-related sites in Germany and the US. These countries differ largely in their risk factors distribution, with a higher smoking prevalence [13–15] and an presumably lower proportion of HPV-positive OCPCs [16] in Germany. Thus, a comparison of these countries will provide further insights about reasons for the strong changes in the epidemiology of OCPC.
Statistical methods – incidence Age standardized incidence rates according to the Segi standard world population were calculated [22]. Rates (per 100,000 persons) for HPV-related OCPC, HPV-unrelated OCPC and HPV-unrelated oral tongue cancer were estimated for Germany and the US for each calendar year between 1999 and 2013. We used “Joinpoint” software for calculation of average annual percentage changes (AAPC) and corresponding 95% confidence intervals as a summary measure for the trend over the whole study period (1999–2013). It is computed as a weighted average of the annual percentage changes (APC) for each segment from the final joinpoint model. We assumed the random errors in the regression model to have constant variance. The regression coefficients were estimated by ordinary least squares [23]. Incidence rates for Germany were calculated using the software CARESS [24]. The US incidence rates were calculated using the software SEER∗Stat [21]. Statistical methods – survival Period analysis [25] was used to derive 5-year relative survival estimates for the time periods 2002–2005, 2006–2009, and 2010–2013 in Germany and the US. Relative survival was calculated as the ratio of the observed survival in the group of OCPC patients divided by the expected survival of a comparable group in the general population. Expected survival was derived according to the Ederer II method [26] from life tables, stratified by age, sex, calendar year, and, in the US, race, as obtained from the German Federal Statistical Office and the Center for Disease Control and Prevention in the US [27]. Age-standardization was performed using weights defined by the International Cancer Survival Standard (age-groups: 15–54, 55–64, 65–74, 75–84, and 85+ years) [28]. Model-based period analysis [29] was employed to test for differences between 5-year relative survival in Germany and US, including adjustment or stratification by age, sex and stage. In the model, the numbers of deaths were modeled as a function of the year of follow-up, country and the adjustment factors by Poisson regression with the logarithm of the person-years at risk as offset. Patients with missing information on stage were excluded in model-based analyses (Germany: 9172 (26%), US: 4235 (14%)). Survival estimates were derived from the models by using standard weights for age-standardization and weights derived from the distribution among all included patients for stage and sex. For subgroups analysis by stage and for stage adjustment, tumors were classified using the TNM Classification of Malignant Tumors (6th and 7th edition) [30,31]. Stage IV tumors were additionally subdivided in stage IV with (M1) and without metastasis (M0). Comparative stage analyses were restricted to the time period 2004–2013, as only for these years TNM information based on the 6th or 7th edition was available in
Patients and methods Data bases For survival analysis, a pooled national dataset from the German cancer survival project was used [17,18]. Briefly, the survival-dataset comprises data from 12 population-based cancer registries encompassing 12 of the 16 federal states and covering 28.3 million inhabitants in 2013 (approx. 35% of the total German population). Regions were selected by their data quality assessed by the proportion of death certificate only (DCO) and autopsy only cases among all registered malignant cancers (Supplementary Table 1). Regions were included if the proportion of DCO cases in the period 2002–2013 was below 13%. Patients age 15 or older diagnosed with a first OCPC tumor in 1997–2013 with a passive mortality follow-up to December 2013 were included in this analysis. Incidence was calculated for the same regions, with the exception of Lower Saxony, where state wide cancer registration did not start before 2003, and which was therefore excluded from incidence analysis. We also restricted the incidence analysis to the period 1999–2013 as three registers started in 1998/99. Patients notified by DCO or autopsy only were excluded. 9
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Table 1 Characteristics of patients with oral cavity or pharynx cancer at HPV related sites and HPV unrelated sites and with HPV-unrelated oral tongue cancer diagnosed between 1997 and 2013 in Germany and the United States. Factor
Level
a
Total Gender Age group
Topography
Stage (2004+)b
Male Female 15–44 45–54 55–64 65–74 75+ Base of tongue (C01) Other parts of tongue (C02) Gum (C03) Floor of mouth (C04) Palate (C05) Other oral cavity (C06) Tonsil (C09) Oropharynx (C10) Hypopharynx (C12 and C13) Other oral & pharynx (C14) I II III IV IV M0 IV M1
HPV-related
HPV-unrelated
HPV-unrelated oral tongue
Germany N (%)
USN (%)
Germany N (%)
USN (%)
Germany N (%)
USN (%)
19,448 (37) 15,401 (79) 4047 (21) 1174 (6) 5296 (27) 7044 (36) 4417 (23) 1517 (8) 4397 (23) 46 (0) na na na na 7808 (40) 7188 (37) na 9 (0) 605 (6) 853 (8) 1624 (15) 7831 (72) 6467 (91) 653 (9)
21,592 (44) 17,420 (81) 4172 (19) 1241 (6) 5565 (26) 7561 (35) 4618 (21) 2607 (12) 9012 (42) 151 (1) na na na na 10,274 (48) 2148 (10) na 7 (0) 746 (6) 968 (7) 2117 (16) 9299 (71) 8546 (94) 559 (6)
25,883 (49) 20,442 (79) 5441 (21) 1613 (6) 6774 (26) 8866 (34) 6120 (24) 2510 (10) na na 2802 (11) 7852 (30) 2514 (10) 2306 (9) na na 8869 (34) 1540 (6) 1798 (15) 1332 (11) 1552 (13) 7545 (62) 5826 (88) 825 (12)
17,408 (36) 11,413 (66) 5995 (34) 541 (3) 2735 (16) 4535 (26) 4709 (27) 4888 (28) na na 2475 (14) 3989 (23) 2180 (13) 3304 (19) na na 4300 (25) 1160 (7) 1817 (23) 1244 (16) 1048 (13) 3762 (48) 3328 (91) 331 (9)
7456 (14) 5170 (69) 2286 (31) 675 (9) 1615 (22) 2270 (30) 1872 (25) 1024 (14) na 7456 (100) na na na na na na na na 944 (27) 615 (17) 598 (17) 1360 (39) 1105 (91) 108 (9)
9861 (20) 5680 (58) 4181 (42) 1172 (12) 1939 (20) 2501 (25) 2085 (21) 2164 (22) na 9861 (100) na na na na na na na na 2430 (45) 931 (17) 753 (14) 1256 (23) 1146 (94) 78 (6)
HPV, Human Papillomavirus; US, United States; na = not applicable. a For total numbers, row percentages by HPV-relation are provided for Germany and the US. b Restricted to patients diagnosed in 2004–2012, 6th and 7th version of the TNM. Patients with missing stage information (Germany: 9172 (26%), US: 4235 (14%)) were excluded. In addition, patients with ICD-O topography code C14 were excluded (Germany: 1070, US: 624), as stage was not defined for these cases in the US. For 1586 stage IV patients in Germany (10% of all stage IV patients) and 329 stage IV patients in the US (2% of all stage IV patients), information on metastasis was missing.
significantly increased between 1999 and 2013 in the US and in Germany, with an average annual increase of 3.3% (95% confidence interval (95%CI) 2.7–3.8) and 2.6% (95%CI 1.7–3.4), respectively (Fig. 1). Incidence was slightly higher in Germany throughout the whole time period. For women, incidence rates for HPV-related sites were comparable in Germany and the US in 1999–2002, but from then rose by 4.4% per year (95%CI 3.3–5.5) in Germany, while remaining stable in the US (Fig. 1). Age specific analysis revealed that this development was mostly driven by the age group 55–64, where incidence rates in Germany more than doubled during the study period (Supplementary Table 3). Divergent trends between the two countries were also observed for HPV-unrelated sites (Fig. 2). In the US, incidence rates annually declined by 2.1% (95% CI −2.8 to −1.4) for men and by 1.7% (95% CI −2.7 to −0.8) for women, while in Germany, incidence rates increased by 0.9% per year (95%CI 0.3–1.4) for men and 2.5% (95%CI 1.7–3.3) for women. Compared to the US, the incidence for HPV-unrelated sites in more recent years was more than twice as high in Germany for men, and about 50% higher for women. In HPV-unrelated oral tongue sites incidence rates slightly increased in both countries and for both sexes, albeit slightly more pronounced in Germany (Supplementary Table 4).
the SEER data. All calculations were carried out with SAS software (version 9.3), using macros developed for period analysis [32] and their model-based extensions [29]. Statistical significance was tested two-sided using a significance level of 0.05. Results Between 1997 and 2013, 19,448 (37%), 25,883 (49%) and 7456 (14%) patients with HPV-related OCPC, HPV-unrelated OCPC and HPVunrelated oral tongue cancer, respectively, were reported to the German registries. In the US, patients had relatively more often HPVrelated OCPC (21,592 (44%)) and HPV-unrelated oral tongue cancer (9861 (20%)) and less often HPV-unrelated OCPC (17,408 (36%)). Table 1 shows the characteristics of OCPC patients by HPV-relation of the cancer site in Germany and the US. For HPV-related sites, sex and stage distributions were comparable in Germany and the US. Around 80% of the patients were male and around 71% of the patients had stage IV disease at time of diagnosis. Patients in the US were slightly more often age 75+. Strong differences were observed with respect to topography with higher proportion of oropharynx cancer and lower proportion of base of tongue cancer in Germany. For HPV-unrelated OCPC, patients in Germany were more often male, were less often older than 75 years, had more often stage IV and floor of mouth and hypopharynx cancer, but less often other oral cavity cancer. For HVP-unrelated oral tongue cancer, patients in Germany were more often male, were less often age 75+, and had more often stage IV. Differences between countries were similar after stratification by sex (Supplementary Table 2).
Survival For HPV-related OCPC, age-standardized 5-year relative survival in 2010–2013 was higher in the US (60.2%) than in Germany (47.5%; Table 2). As the survival advantage was only observed in men (+16.7% units compared to −0.2% units in women), further subgroup analyses were conducted after stratification by sex. For men, survival was higher in the US in all age, topography and stage groups (all p < .0001) except in patients with distant metastasis. In adjusted models, differences were largest in men aged 15–54 years (+27.8% units) and for stage IV M0 (+26.4% units). In women, overall age-standardized 5-year relative
Incidence Age standardized incidence rates for HPV-related sites in men 10
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Fig. 1. Age-standardized incidence and average annual percentage change (AAPC) of oral cavity and pharynx cancer at HPV-related sites for male and female patients in Germany and the United States in 1999–2013.
improvements for HPV-related OCPC were comparable across age groups (+7.1 to +9.4% units) and again larger than in Germany (−9.5 to 2.8% units), leading to an increasing survival benefit for women aged 15–54 and 55–64 in the US that was, however, not significant. For elderly women, survival was significantly higher in Germany in 2002–2005 (+20.8% units) but decreased over time and was only 4.2% units higher in Germany in 2010–2013.
survival in Germany and the US were comparable (53.5 and 53.3%). Significant differences in adjusted models were only observed for stage I-III and for oropharynx cancer showing higher survival in Germany (+11.3 and +12.0% units; p = .0124 and p = .0212). Five-year age-standardized relative survival for HPV-unrelated OCPC was comparable in Germany and the US (44.0 and 45.6%; Table 2). After adjustment for stage, survival was significantly higher in Germany, overall and for males (both: +2.6% units). For HPV-unrelated oral tongue cancer, 5-year age-standardized relative survival was higher in the US than in Germany overall (+6.8% units; Table 2) and for men (+7.0% units) and women (+4.5% units). However, after adjustment for stage, survival estimates in Germany and the US were comparable with differences smaller than 1.0% unit. Country comparison of 5-year relative survival in 2002–2005 and 2006–2009 as well as survival trends between 2002–2005 and 2010–2013 are shown in Fig. 3 and Supplementary Table 5. In Germany, modest improvements in 5-year age-standardized relative survival by 2.6, 3.6 and 2.6% units were observed for HPV-related OCPC, HPV-unrelated OCPC and oral tongue cancer. In contrast, in the US, a steep rise in survival rates by 13.3% units was seen for HPV-related OCPC, whereas survival increased by about 5% units for the other sites. For HPV-related sites in men in the US, the improvements were more pronounced in older age (+8.4, +12.1 and +20.5% units for patients aged 15–54, 55–64 and 65+ years). In Germany, increases were generally smaller and no consistent age pattern was observed. Consequently, the survival difference between Germany and the US became more distinct in recent years. While in men age 15–54 years survival was already 24.3% units higher in the US in 2002–2005, men aged 65 years and older had comparable survival in 2002–2005 but much higher survival in the US in 2010–13 (+20.5% units). In US women,
Sensitivity analyses In sensitivity analyses, we restricted all analyses for the US to the white population (40.081 incident cases). The characteristics of white OCPC patients (Supplementary Table 7) were comparable with the characteristics of the total group. After restricting to whites, incidence of HPV-related cancers was higher in recent years, particularly among males, while no substantial changes were observed at HPV-unrelated sites (Supplementary Figs. 1 and 2). Survival for white OCPC patients in the US was higher compared to results for all US patients in all subgroups (Supplementary Table 7, Supplementary Fig. 3). Consequently, the previously described survival advantage for HPV-related cancers in the US compared to Germany was larger. Discussion The present study provides a comprehensive comparison of OCPC incidence and survival in Germany and the US showing strong differences across countries. For HPV-related sites in men, age-standardized incidence rates were largely comparable between both countries, with increasing trends in the US and Germany. Among women, an increasing gap between the US and Germany was observed: incidence rates 11
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Fig. 2. Age-standardized incidence and average annual percentage change (AAPC) of oral cavity and pharynx cancer at HPV-unrelated sites for male and female patients in Germany and the United States in 1999–2013.
strong geographical variations [38]. A recent international study reported substantially lower HPV-prevalence in Europe (31.1%) compared to North-America (59.3%) for the years 2002–2005/06 [16]. Therefore, it can be assumed that, compared to Germany, the proportion of HPV-positive tumors was higher in the US in our study period. It should be mentioned that these findings do not automatically indicate a higher HPV-prevalence in the US population, as the lower impact of smoking and alcohol would at least partly explain the higher relative contribution of HPV in the US. It is further known that the prevalence of oral oncogenic HPV-infection has been described to be 3-times higher among men than women. This finding is consistent with the higher risk for HPV-positive oropharyngeal cancer among men in the US [39]. Recent findings from NHANES, the U.S. National Health and Nutritional Examination Survey, indicate that the higher prevalence of oral oncogenic HPV-infection among men arises from a higher number of lifetime oral sexual partners and stronger associations between oral sex and infection among men [40,41]. HPV-associated OCPC trends might correspondingly be driven predominantly by changes in sexual behavior in men. Strong country-differences in smoking, alcohol consumption and proportion of HPV-positive tumors might also explain survival differences. Our results revealed significant differences in survival after OCPC at HPV-related sites in 2010–2013, with relative survival rates in the US being up to 27.4% units higher compared to Germany.. Strong differences were only observed in men and were largest in younger patients and stage IV M0 disease. While in men below the age of 65 years, these differences were, to a lesser extent, already present in 2002–2005, they became apparent in older men only in more recent years. It has been shown that smoking and HPV-prevalence influence the
significantly increased in Germany while remaining stable in the US. For HPV-unrelated sites, incidence was much higher in Germany than in the US, particularly in men, and the trends were divergent with increasing rates in Germany but decreasing ones in the US. For HPVunrelated oral tongue cancer, no pronounced differences were observed between the two countries. Our results on OCPC incidence are consistent with previous studies from the US [7–9] and from smaller areas in Germany [33,34]. Incidence differences between Germany and the US are likely to result from different risk factor distributions. In the US, tobacco use decreased among men between 1992/93 and 2010/11 from 26.5% to 20.1% [13,15]. In Germany, smoking prevalence for men also declined, but still remains on a higher level (1992: 36.8%, 2013: 29.0%) [14]. For women, smoking prevalence was comparable in Germany [14] and the US [13,15] in 1992 (21.8% and 21.5%). While it decreased steadily in the US (2010/11: 15.5%), it initially increased in Germany to 32% in 2003 and afterwards dropped to 20.3% in 2013. These patterns are consistent with the observed overall higher OCPC incidence in German males, especially at HPV-unrelated sites and with a comparable OCPC incidence for females between 1999 and 2002 followed by an increasing incidence in Germany in later years. In addition to differences in smoking prevalence, higher alcohol consumption in Germany is another factor that might explain higher OCPC incidence [35]. Although smoking and alcohol consumption also contribute to the disease burden of HPV-related OCPC, there is convincing evidence that HPV-infection is an independent etiologic factor for these tumors. Its importance seems to be growing, as several observations indicate that HPV-prevalence in OCPC has increased over time in different populations [33,36,37]. In a recent meta-analysis, it was estimated that 39.8% of oropharyngeal cancers could be attributed to HPV16-infection with 12
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Table 2 Country comparisons of 5-year relative survival for male and female patients with oral cavity or pharynx cancer in Germany and the United States in 2010–2013. HPV-relation
Sex
Subgroup
Germany
Country comparison: Germany – US
US
N
RS
SE
N
RS
SE
Diffb
Adj. diffc
Adj. Pc
Overalla Overalla 15–54 years 55–64 years 65+ years Base of tonguea Tonsila Oropharynxa Stage I/II/IIIa Stage IV M0a Stage IV M1a Overalla 15–54 years 55–64 years 65+ years Base of tonguea Tonsila Oropharynxa Stage I/II/IIIa Stage IV M0a Stage IV M1a
12,603 9923 3206 3606 3111 2336 3863 3707 2147 4746 501 2680 747 888 1045 635 1207 831 660 1182 98
47.5 45.4 49.0 50.3 44.6 44.5 54.6 36.3 62.2 44.4 16.8 53.5 61.8 61.6 48.9 47.9 64.4 43.3 74.8 47.9 ∼
1.1 1.4 1.4 1.4 1.6 2.6 2.4 2.2 3.2 2.3 4.1 2.0 2.8 2.6 2.6 3.6 3.2 3.9 5.0 3.0 ∼
13,621 11,199 3339 4482 3378 4775 5327 1026 2740 6732 406 2422 644 760 1018 953 1149 302 776 1170 107
60.2 62.1 76.4 71.3 57.3 60.1 67.5 45.5 69.6 62.6 14.1 53.3 69.1 65.1 44.7 50.7 60.6 31.7 59.3 55.6 ∼
1.0 1.1 1.2 1.1 1.6 1.6 1.9 3.7 2.3 1.6 3.0 1.9 2.9 2.8 2.8 3.0 2.8 4.6 3.2 2.9 ∼
−12.7 −16.7 −27.4 −21.0 −12.7 −15.6 −12.9 −9.2 −7.4 −18.2 2.7 0.2 −7.3 −3.5 4.2 −2.8 3.8 11.6 15.5 −7.7 ∼
−17.8 −22.1 −27.8 −22.1 −11.5 −16.9 −19.0 −9.0 −12.3 −26.4 −3.2 0.8 −8.4 −1.1 5.8 −1.1 2.6 12.0 11.3 −5.4 −0.1
< .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 .2686 .7251 .0523 .7778 .1198 .7963 .4735 .0212 .0124 .0986 .9816
Both Male Female
Overalla Overalla Overalla
16,154 12,526 3628
44.0 40.8 53.4
0.9 1.1 1.6
9049 5959 3090
45.6 42.6 51.4
1.0 1.2 1.7
−1.6 −1.8 2.0
2.6 2.6 2.8
.0106 .0365 .1972
HPV-unrelated oral tongue Both Male Female
Overalla Overalla Overalla
4696 3209 1487
56.9 53.0 64.2
1.6 2.2 2.4
5829 3347 2482
63.7 60.0 68.7
1.3 1.8 1.9
−6.8 −7.0 −4.5
−0.3 −0.1 −0.9
.8343 .9639 .7443
HPV-related Both Male
Female
HPV-unrelated
HPV, Human Papillomavirus; US, United States; N, total number of cases; RS, 5-year relative survival; SE, standard errors of the estimate of RS; Diff, Difference between RS estimates; Adj., adjusted; na, not applicable; ∼, standard error of the RS period analysis was larger than 5.0. a Age-adjusted using five age groups (15–44, 45–54, 55–64, 65–74, and 75+ years). b Differences of RS between Germany and the US based on period analysis. c Differences of RS between Germany and the US and corresponding p-value based on model based period analysis including adjustment/stratification for age and stage (significant differences in bold).
only minor, while for oral tongue tumors they diminished when adjusting for tumor stage. In contrast to survival at HPV-related sites, we could not observe different patterns in country comparisons by sex or by calendar period for HPV-unrelated OCPC supporting the hypothesis of different etiologic fractions in both countries being responsible for the observed differences in survival of HPC-related sites. Our study provides the first comparison of OCPC survival by HPVrelation of the cancer site in Germany and the US, two countries with very different smoking and HPV-prevalence. Strengths include the use of two large high quality datasets that allowed describing survival differences on the population-level and conducting detailed subgroup analyses by age, sex, topography and stage. Furthermore, we were able to investigate both incidence and survival for the same underlying population. A limitation of the study is that stage information for the US data was only available since 2004 and, thus, could not be used in trend analyses. Furthermore, a considerable number of patients had to be excluded in stage-adjusted/stratified analyses due to missing information. Another limitation is the lack of detailed treatment data, especially for Germany. Potential differences in the registration practice and coding of OCPC in Germany and the US are further limitations. The distribution of the topographies was different in Germany and the US. As we were not able to decide whether this was caused by differences in coding practice or reflects a real incidence difference, we did not include topography as adjustment factor in the models. However, higher survival rates in the US for HPV-related sites were observed for all topographies. Another issue is the use of different multiple primary rules in Germany and the
prognosis of OCPC. Smoking during therapy decreases the efficacy of radiotherapy in head and neck cancers [42]. Furthermore, smoking at diagnosis and during therapy has been reported to be associated with higher mortality after oropharyngeal cancer, independent of the tumor HPV-status [43]. Patients with HPV-positive tumors have higher response rates to chemoradiotherapy [44] and are consistently demonstrated to have a better prognosis than patients with HPV-negative tumors [43,45,46]. Thus, the differences in smoking and HPV-positive tumor prevalence might at least partly explain the observed survival benefit for HPV-related OCPC in the US compared to Germany. The finding that these differences were most pronounced for younger men supports this explanation, as HPV-positive tumors are more prevalent in this group [16,36]. Furthermore, the observation that the survival advantage in the US was already present in young men in 2002–2005 but emerged more recently for patients age 65+ is consistent with the delayed incidence increase among elderly men in the US, which was thought to be explained by a later increase in the proportion of HPVpositive tumors in this age group [9]. To explain cross-country differences in survival, differences in diagnosis and treatment, potentially reflected in clinical guidelines, have to be considered. For head and neck cancer. it has previously been noted that NCCN guidelines (US) would be more comprehensive and provide greater clarity with respect to day-to-day clinical practice than ESMO guidelines (Europe) [47]. However, survival differences caused by differences in treatment can be expected to be present in both men and women and, thus, would not explain our findings. For HPV-unrelated OCPC, survival differences across countries were 13
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
OCPC by HPV-relation of site
Five-year relative survival [%]
HPV-related, GER HPV-related, US
HPV-unrelated, GER HPV-unrelated, US
be done with caution for studies that do not consider the HPV-status. Finally, close monitoring of the epidemiology of OCPC in each country is needed and should be performed after stratification of OCPC sites according to potential HPV association.
Oral tongue, GER Oral tongue, US
80 75 70 65 60 55 50 45 40 35 30
Conflict of interest The authors declare that they have no conflict of interest. Acknowledgement 2002-2005
2006-2009
Members of the GEKID Cancer Survival Working Group: Karla Geiss, Martin Meyer (Cancer Registry of Bavaria), Andrea Eberle, Sabine Luttmann (Cancer Registry of Bremen), Roland Stabenow (Cancer Registry of Berlin and the New Federal States), Stefan Hentschel, Alice Nennecke (Hamburg Cancer Registry), Joachim Kieschke, Eunice Sirri (Cancer Registry of Lower Saxony), Bernd Holleczek (Saarland Cancer Registry), Katharina Emrich (Cancer Registry of Rhineland-Palatinate), Hiltraud Kajüter, Volkmar Mattauch (Cancer Registry of North RhineWestphalia), Alexander Katalinic, Nora Eisemann (Cancer Registry of Schleswig-Holstein), Klaus Kraywinkel (Robert Koch Institute, Berlin), Hermann Brenner, Lina Jansen and Janick Weberpals (German Cancer Research Center).
2010-2013
Calendar period OCPC at HPV-related sites in men
Five-year relative survival [%]
15-54 years, GER 15-54 years, US
55-64 years, GER 55-64 years, US
65+ years, GER 65+ years, US
2006-2009
2010-2013
80 75 70 65 60 55 50 45 40 35 30
Funding 2002-2005
This work was supported by the German Cancer Aid (Deutsche Krebshilfe) [grant number 108257 and 110446].
Calendar period
OCPC at HPV-related sites in women
Five-year relative survival [%]
15-54 years, GER 15-54 years, US
55-64 years, GER 55-64 years, US
Appendix A. Supplementary material
65+ years, GER 65+ years, US
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.oraloncology.2017.11. 015.
80 75 70 65 60 55 50 45 40 35 30
References
2002-2005
2006-2009
[1] Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, et al. GLOBOCAN 2012 v1.0, cancer incidence and mortality worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: < http://globocan.iarc.fr > [accessed 20/07/2016]. [2] International Agency for Research on Cancer, editor. Human papillomaviruses, IARC monographs on the evaluation of carcinogenic risks to humans. Lyon, France: World Health Organization; 2007. p 670. [3] Gillison ML, Chaturvedi AK, Anderson WF, Fakhry C. Epidemiology of human papillomavirus-positive head and neck squamous cell carcinoma. J Clin Oncol 2015;33:3235–42. [4] Gillison ML, D'Souza G, Westra W, Sugar E, Xiao W, Begum S, et al. Distinct risk factor profiles for human papillomavirus type 16-positive and human papillomavirus type 16-negative head and neck cancers. J Natl Cancer Inst 2008;100:407–20. [5] Day GL, Blot WJ, Austin DF, Bernstein L, Greenberg RS, Preston-Martin S, et al. Racial differences in risk of oral and pharyngeal cancer: alcohol, tobacco, and other determinants. J Natl Cancer Inst 1993;85:465–73. [6] Mehanna H, Beech T, Nicholson T, El-Hariry I, McConkey C, Paleri V, et al. Prevalence of human papillomavirus in oropharyngeal and nonoropharyngeal head and neck cancer – systematic review and meta-analysis of trends by time and region. Head Neck 2013;35:747–55. [7] Chaturvedi AK, Anderson WF, Lortet-Tieulent J, Curado MP, Ferlay J, Franceschi S, et al. Worldwide trends in incidence rates for oral cavity and oropharyngeal cancers. J Clin Oncol 2013;31:4550–9. [8] Chaturvedi AK, Engels EA, Anderson WF, Gillison ML. Incidence trends for human papillomavirus-related and -unrelated oral squamous cell carcinomas in the United States. J Clin Oncol 2008;26:612–9. [9] Zumsteg ZS, Cook-Wiens G, Yoshida E, Shiao SL, Lee NY, Mita A, et al. Incidence of oropharyngeal cancer among elderly patients in the United States. JAMA Oncol 2016;2:1617–23. [10] Gatta G, Botta L, Sanchez MJ, Anderson LA, Pierannunzio D, Licitra L, et al. Prognoses and improvement for head and neck cancers diagnosed in Europe in early 2000s: the EUROCARE-5 population-based study. Eur J Cancer 2015. [11] O'Rorke MA, Ellison MV, Murray LJ, Moran M, James J, Anderson LA. Human papillomavirus related head and neck cancer survival: a systematic review and meta-analysis. Oral Oncol 2012;48:1191–201. [12] Gillison ML, Restighini C. Anticipation of the impact of human papillomavirus on clinical decision making for the head and neck cancer patient. Hematol Oncol Clin
2010-2013
Calendar period Fig. 3. Age-standardized overall relative survival in 2002–2005, 2006–2009 and 2010–2013 in Germany and the United States for oral cavity and pharynx cancer at HPVrelated sites, HPV-unrelated sites and at HPV-unrelated oral tongue and for oral cavity and pharynx cancer at HPV-related sites by sex and age.
US. However, it has been shown that the difference between these two rules has only minor effects for OCPCs [48]. Differences in the completeness of registration might also bias the comparison. In both countries, the proportion of DCO cases cannot reliably be incorporated in our study, as cases were selected by ICD-O-3 histology codes that are very rarely specified on death certificates. However, in previous studies it has been shown that DCO proportions are generally higher in Germany than in the US [18], potentially leading to an overestimation of survival and underestimation of incidence in Germany. Thus, the reported survival advantage for HPV-related OCPC in the US might even be larger. Conclusion Due to the observed trends that indicate differences in patient characteristics with respect to HPV prevalence across countries, country comparisons should be stratified by HPV status or at least by HPV-associations of the cancer site. Furthermore, translation of evidence, e.g. with respect to prognosis and treatment, across countries should only 14
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
[31] Sobin LH, Gospodarowicz MK, Wittekind C. TNM classification of malignant tumors, 7th ed. Oxford, UK: Wiley-Blackwell; 2009. [32] Brenner H, Gefeller O, Hakulinen T. A computer program for period analysis of cancer patient survival. Eur J Cancer 2002;38:690–5. [33] Tinhofer I, Johrens K, Keilholz U, Kaufmann A, Lehmann A, Weichert W, et al. Contribution of human papilloma virus to the incidence of squamous cell carcinoma of the head and neck in a European population with high smoking prevalence. Eur J Cancer 2015;51:514–21. [34] Bayer O, Kruger M, Koutsimpelas D, Emrich K, Ressing M, Zeissig SR, et al. Changes in incidence and mortality trends of head and neck cancer in Rhineland-palatinate, 2000–2009. Laryngorhinootologie 2015;94:451–8. [35] World Health Organization. Global status report on alcohol and health 2014. World Health Organization; 2014. [36] Castellsague X, Alemany L, Quer M, Halec G, Quiros B, Tous S, et al. HPV involvement in head and neck cancers: comprehensive assessment of biomarkers in 3680 patients. J Natl Cancer Inst 2016;108. djv403. [37] Chaturvedi AK, Engels EA, Pfeiffer RM, Hernandez BY, Xiao W, Kim E, et al. Human papillomavirus and rising oropharyngeal cancer incidence in the United States. J Clin Oncol 2011;29:4294–301. [38] Ndiaye C, Mena M, Alemany L, Arbyn M, Castellsague X, Laporte L, et al. HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis. Lancet Oncol 2014;15:1319–31. [39] Gillison ML, Broutian T, Pickard RK, Tong ZY, Xiao W, Kahle L, et al. Prevalence of oral HPV infection in the United States, 2009–2010. JAMA 2012;307:693–703. [40] D'Souza G, Cullen K, Bowie J, Thorpe R, Fakhry C. Differences in oral sexual behaviors by gender, age, and race explain observed differences in prevalence of oral human papillomavirus infection. PLoS ONE 2014;9:e86023. [41] Chaturvedi AK, Graubard BI, Broutian T, Pickard RK, Tong ZY, Xiao W, et al. NHANES 2009–2012 findings: association of sexual behaviors with higher prevalence of oral oncogenic human papillomavirus infections in U.S. men. Can Res 2015;75:2468–77. [42] Browman GP, Wong G, Hodson I, Sathya J, Russell R, McAlpine L, et al. Influence of cigarette smoking on the efficacy of radiation therapy in head and neck cancer. N Engl J Med 1993;328:159–63. [43] Gillison ML, Zhang Q, Jordan R, Xiao W, Westra WH, Trotti A, et al. Tobacco smoking and increased risk of death and progression for patients with p16-positive and p16-negative oropharyngeal cancer. J Clin Oncol 2012;30:2102–11. [44] Fakhry C, Westra WH, Li S, Cmelak A, Ridge JA, Pinto H, et al. Improved survival of patients with human papillomavirus-positive head and neck squamous cell carcinoma in a prospective clinical trial. J Natl Cancer Inst 2008;100:261–9. [45] Goodman MT, Saraiya M, Thompson TD, Steinau M, Hernandez BY, Lynch CF, et al. Human papillomavirus genotype and oropharynx cancer survival in the United States of America. Eur J Cancer 2015;51:2759–67. [46] D'Souza G, Anantharaman D, Gheit T, Abedi-Ardekani B, Beachler DC, Conway DI, et al. Effect of HPV on head and neck cancer patient survival, by region and tumor site: a comparison of 1362 cases across three continents. Oral Oncol 2016;62:20–7. [47] D'Cruz A, Lin T, Anand AK, Atmakusuma D, Calaguas MJ, Chitapanarux I, et al. Consensus recommendations for management of head and neck cancer in Asian countries: a review of international guidelines. Oral Oncol 2013;49:872–7. [48] Weir HK, Johnson CJ, Thompson TD. The effect of multiple primary rules on population-based cancer survival. Cancer Causes Control 2013;24:1231–42.
North Am 2015;29:1045–60. [13] US Department of Commerce, Census Bureau. National Cancer Institute-sponsored Tobacco Use Supplement to the Current Population Survey (2010–11); 2012. < http://appliedresearch.cancer.gov/tus-cps/results/data1011/table 1. html > [accessed 28/07/2016]. [14] Microcensus – health questions. Federal Statistical Office, Branch Bonn [accessed 06/04/2017]. [15] Jemal A, Thun M, Yu XQ, Hartman AM, Cokkinides V, Center MM, et al. Changes in smoking prevalence among U.S. adults by state and region: estimates from the Tobacco Use Supplement to the Current Population Survey, 1992–2007. BMC Public Health 2011;11:512. [16] Anantharaman D, Abedi-Ardekani B, Beachler DC, Gheit T, Olshan AF, Wisniewski K, et al. Geographic heterogeneity in the prevalence of human papillomavirus in head and neck cancer. Int J Cancer 2017;140:1968–75. [17] Hiripi E, Gondos A, Emrich K, Holleczek B, Katalinic A, Luttmann S, et al. Survival from common and rare cancers in Germany in the early 21st century. Ann Oncol 2011;23:472–9. [18] Jansen L, Castro FA, Gondos A, Krilaviciute A, Barnes B, Eberle A, et al. Recent cancer survival in Germany: an analysis of common and less common cancers. Int J Cancer 2015;136:2649–58. [19] Brown LM, Check DP, Devesa SS. Oral cavity and pharynx cancer incidence trends by subsite in the United States: changing gender patterns. J Oncol 2012;2012:649498. [20] Fritz APC, Jack A, Shanmugaratnam K, Sobin LH, Parkin DM, et al. International classification of diseases for oncology. 3rd ed. Geneva: World Health Organization; 2000. [21] Surveillance, Epidemiology, and End Results (SEER) Program Research Data (1973–2013). National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2016, based on the November 2015 submission. < www.seer.cancer.gov > . [22] Bray F, Ferlay J. Age-standardisation. In: Forman DFB, Brewster DH, Gombe Mbalawa C, Kohler B, Piñeros M, Steliarova-Foucher E, editors. Cancer incidence in five continents Vol X. Lyon: IARC; 2007 [chapter 7]. [23] Statistical research and applications branch NCI. Joinpoint regression program. 4.0. 4. ed2013. [24] OFFIS e.V. CARESS. Analytisches Informationssystem für Krebsregisterdaten. 8.8. 12.6 ed. Oldenburg, Germany; 2016. [25] Brenner H, Gefeller O, Hakulinen T. Period analysis for ‘up-to-date' cancer survival data: theory, empirical evaluation, computational realisation and applications. Eur J Cancer 2004;40:326–35. [26] Ederer F, Heise H. Instructions to IBM 650 programmers in processing survival computations. Methodological note no. 10. End results evaluation section. Bethesda, MD: National Cancer Institute; 1959. [27] Center for Disease Control and Prevention Publications and Information Products Life Tables. < http://www.cdc.gov/nchs/products/life_tables.htm > [accessed 01/ 07/2016]. [28] Corazziari I, Quinn M, Capocaccia R. Standard cancer patient population for age standardising survival ratios. Eur J Cancer 2004;40:2307–16. [29] Brenner H, Hakulinen T. Up-to-date and precise estimates of cancer patient survival: model-based period analysis. Am J Epidemiol 2006;164:689–96. [30] Sobin LH, Wittekind C. TNM classification of malignant tumours. 6th ed. New York: Wiley-Liss; 2002.
15
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
Differences in incidence and survival of oral cavity and pharyngeal cancers between Germany and the United States depend on the HPV-association of the cancer site
T
⁎
L. Jansena, , N. Buttmann-Schweigerb, S. Listlc,d, M. Ressinge,f, B. Holleczekg, A. Katalinich, S. Luttmanni, K. Kraywinkelb, H. Brennera,j,k, for the GEKID Cancer Survival Working Group1 a
Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany German Centre for Cancer Registry Data (ZfKD), Department of Epidemiology and Health Monitoring, Robert Koch-Institute, Berlin, Germany Department of Conservative Dentistry, University Hospital Heidelberg, Heidelberg, Germany d Radboud University Medical Center, Radboud Institute for Health Sciences, Department of Dentistry – Quality and Safety of Oral Health Care, Nijmegen, The Netherlands e Cancer Registry Rhineland-Palatinate, Mainz, Germany f Institute of Medical Biostatistics, Epidemiology and Informatics, University Medical Center, Johannes Gutenberg University Mainz, Mainz, Germany g Saarland Cancer Registry, Saarbrücken, Germany h Cancer Registry of Schleswig-Holstein, Lübeck, Germany i Cancer Registry of Bremen, Leibniz-Institute for Prevention Research and Epidemiology – BIPS, Bremen, Germany j Division of Preventive Oncology, German Cancer Research Center (DKFZ), and National Center for Tumor Diseases (NCT), Heidelberg, Germany k German Cancer Consortium (DKTK), German Cancer Research Center, Heidelberg, Germany b c
A R T I C L E I N F O
A B S T R A C T
Keywords: Oral cancer Pharynx cancer HPV Survival Incidence Trend Germany United States
Introduction: The epidemiology of squamous cell oral cavity and pharyngeal cancers (OCPC) has changed rapidly during the last years, possibly due to an increase of human papilloma virus (HPV) positive tumors and successes in tobacco prevention. Here, we compare incidence and survival of OCPC by HPV-relation of the site in Germany and the United States (US). Materials and methods: Age-standardized and age-specific incidence and 5-year relative survival was estimated using data from population-based cancer registries in Germany and the US Surveillance Epidemiology and End Results (SEER) 13 database. Incidence was estimated for each year between 1999 and 2013. Relative survival for 2002–2005, 2006–2009, and 2010–2013 was estimated using period analysis. Results: The datasets included 52,787 and 48,861 cases with OCPC diagnosis between 1997 and 2013 in Germany and the US. Incidence was much higher in Germany compared to the US for HPV-unrelated OCPC and more recently also for HPV-related OCPC in women. Five-year relative survival differences between Germany and the US were small for HPV-unrelated OCPC. For HPV-related OCPC, men had higher survival in the US (62.1%) than in Germany (45.4%) in 2010–2013. These differences increased over time and were largest in younger patients and stage IV disease without metastasis. In contrast, women had comparable survival for HPVrelated OCPC in both countries. Conclusions: Strong survival differences between Germany and the US were observed for HPV-related OCPC in men, which might be explained by differences in HPV-attributable proportions. Close monitoring of the epidemiology of OCPC in each country is needed.
Introduction Oral cavity and pharyngeal cancers (OCPC) include cancers of the lip, oral cavity, and pharynx. Together, this group of cancers accounts
for over 500,000 new cancer cases per year worldwide [1]. In 2007, the WHO concluded that human papillomavirus (HPV) type 16 is a cause of some subtypes of OCPC [2]. Many epidemiological and molecular studies have meanwhile provided evidence for this causal link, particularly
Abbreviations: OCPC, squamous cell oral cavity and pharyngeal cancers; HPV, human papillomavirus; US, United States; SEER, Surveillance, Epidemiology, and End Results; DCO, death certificate only; AAPC, average annual percentage changes; 95% CI, 95% confidence interval ⁎ Corresponding author at: Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 581, Heidelberg 69120, Germany. E-mail address: [email protected] (L. Jansen). 1 Members of the GEKID Cancer Survival Working Group are listed in the acknowledgments. https://doi.org/10.1016/j.oraloncology.2017.11.015 Received 21 July 2017; Received in revised form 9 November 2017; Accepted 14 November 2017 1368-8375/ © 2017 Elsevier Ltd. All rights reserved.
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
We selected and classified OCPC cancers based on the anatomic site and etiologic relationship with HPV following two publications [8,19]. The classification was based on the International Classification of Disease for Oncology version-3 (ICD-0-3) [20]. Cancer with morphology codes 8050–8076, 8078, 8083, 8084 and 8094 (squamous cell carcinoma) with malignant behavior (behavior code 3) of the oral cavity (topography codes C01.9-C06.9) and pharynx (codes C09.0-C10.9, C12.9-C14.8) (excluding lip, salivary glands, and nasopharynx) were included and further classified in three groups: HPV-related OCPC, HPV-unrelated OCPC, and HPV-unrelated oral tongue (C02.0-C02.3, C02.8-C02.9) (Supplementary Material). Data from the Surveillance, Epidemiology, and End Results (SEER) 13 database was used to estimate incidence and survival for the US [21]. SEER-13 covers about 41 million inhabitants (13.4% of the total US population). The same selection and classification criteria as for the German dataset were applied for the US. In sensitivity analyses, analyses were conducted after restricting the US data to Whites.
for squamous cell cancers of the oropharynx, the tonsils and parts of the tongue [3]. Consequently, a paradigm shift in the understanding of OCPC has occurred. HPV-positive and HPV-negative OCPCs have different risk factor profiles [4]. Tobacco use and alcohol consumption are considered the primary risk factors for HPV-negative OCPCs. In contrast, HPV-positive OCPCs are associated with several measures of sexual behavior and marijuana use, which have not been found to be associated with HPVnegative OCPCs. It has been estimated that the combined effects of tobacco use and alcohol consumption are responsible for 70–80% of OCPCs in the United States (US) [5], while HPV was reported to be associated with 46% of oropharyngeal carcinomas [6]. However, the attributable fractions vary strongly by geographical location and calendar time due to changes in the prevalence of HPV-infection, tobacco use and alcohol consumption. In many high-income countries, the incidence of OCPC at HPV-unrelated sites has declined, presumably because of reduced tobacco and alcohol use. In contrast, incidence for HPV-related oropharyngeal cancers has risen strongly, specifically among males [7]. While in earlier studies increases in incidence for HPV-related sites were mainly observed among ages 40–59 years [8], they have recently also been observed in patients aged 65+ years with similar magnitude [9], highlighting the rapidly changing epidemiology of OCPC. Changes in survival patterns after OCPC have also been reported in various countries. In Europe, 5-year relative survival increased between 1999–2001 and 2005–2007 from 42.3 to 47.4 for tongue and lingual tonsil and from 36.7 to 40.1 for oropharynx and tonsil cancer. For oral cavity cancer, the improvement in survival was less pronounced (45.6–48.1) [10]. In the US, survival for oropharyngeal cancer rose considerably in the last years, while it was stable over time for HPVunrelated OCPC sites [8]. These differences have been explained by the increasing proportion of HPV-positive OCPCs together with the more favorable prognosis of HPV-positive OCPC [11]. Thus, the epidemiology of OCPC is changing strongly over time and will lead to modifications in OCPC management [12]. The magnitude of these changes will depend on changes in the prevalence of the main risk factors tobacco use and HPV. Here, we provide a comprehensive comparison of incidence and survival of OCPC at HPV-related and non-related sites in Germany and the US. These countries differ largely in their risk factors distribution, with a higher smoking prevalence [13–15] and an presumably lower proportion of HPV-positive OCPCs [16] in Germany. Thus, a comparison of these countries will provide further insights about reasons for the strong changes in the epidemiology of OCPC.
Statistical methods – incidence Age standardized incidence rates according to the Segi standard world population were calculated [22]. Rates (per 100,000 persons) for HPV-related OCPC, HPV-unrelated OCPC and HPV-unrelated oral tongue cancer were estimated for Germany and the US for each calendar year between 1999 and 2013. We used “Joinpoint” software for calculation of average annual percentage changes (AAPC) and corresponding 95% confidence intervals as a summary measure for the trend over the whole study period (1999–2013). It is computed as a weighted average of the annual percentage changes (APC) for each segment from the final joinpoint model. We assumed the random errors in the regression model to have constant variance. The regression coefficients were estimated by ordinary least squares [23]. Incidence rates for Germany were calculated using the software CARESS [24]. The US incidence rates were calculated using the software SEER∗Stat [21]. Statistical methods – survival Period analysis [25] was used to derive 5-year relative survival estimates for the time periods 2002–2005, 2006–2009, and 2010–2013 in Germany and the US. Relative survival was calculated as the ratio of the observed survival in the group of OCPC patients divided by the expected survival of a comparable group in the general population. Expected survival was derived according to the Ederer II method [26] from life tables, stratified by age, sex, calendar year, and, in the US, race, as obtained from the German Federal Statistical Office and the Center for Disease Control and Prevention in the US [27]. Age-standardization was performed using weights defined by the International Cancer Survival Standard (age-groups: 15–54, 55–64, 65–74, 75–84, and 85+ years) [28]. Model-based period analysis [29] was employed to test for differences between 5-year relative survival in Germany and US, including adjustment or stratification by age, sex and stage. In the model, the numbers of deaths were modeled as a function of the year of follow-up, country and the adjustment factors by Poisson regression with the logarithm of the person-years at risk as offset. Patients with missing information on stage were excluded in model-based analyses (Germany: 9172 (26%), US: 4235 (14%)). Survival estimates were derived from the models by using standard weights for age-standardization and weights derived from the distribution among all included patients for stage and sex. For subgroups analysis by stage and for stage adjustment, tumors were classified using the TNM Classification of Malignant Tumors (6th and 7th edition) [30,31]. Stage IV tumors were additionally subdivided in stage IV with (M1) and without metastasis (M0). Comparative stage analyses were restricted to the time period 2004–2013, as only for these years TNM information based on the 6th or 7th edition was available in
Patients and methods Data bases For survival analysis, a pooled national dataset from the German cancer survival project was used [17,18]. Briefly, the survival-dataset comprises data from 12 population-based cancer registries encompassing 12 of the 16 federal states and covering 28.3 million inhabitants in 2013 (approx. 35% of the total German population). Regions were selected by their data quality assessed by the proportion of death certificate only (DCO) and autopsy only cases among all registered malignant cancers (Supplementary Table 1). Regions were included if the proportion of DCO cases in the period 2002–2013 was below 13%. Patients age 15 or older diagnosed with a first OCPC tumor in 1997–2013 with a passive mortality follow-up to December 2013 were included in this analysis. Incidence was calculated for the same regions, with the exception of Lower Saxony, where state wide cancer registration did not start before 2003, and which was therefore excluded from incidence analysis. We also restricted the incidence analysis to the period 1999–2013 as three registers started in 1998/99. Patients notified by DCO or autopsy only were excluded. 9
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Table 1 Characteristics of patients with oral cavity or pharynx cancer at HPV related sites and HPV unrelated sites and with HPV-unrelated oral tongue cancer diagnosed between 1997 and 2013 in Germany and the United States. Factor
Level
a
Total Gender Age group
Topography
Stage (2004+)b
Male Female 15–44 45–54 55–64 65–74 75+ Base of tongue (C01) Other parts of tongue (C02) Gum (C03) Floor of mouth (C04) Palate (C05) Other oral cavity (C06) Tonsil (C09) Oropharynx (C10) Hypopharynx (C12 and C13) Other oral & pharynx (C14) I II III IV IV M0 IV M1
HPV-related
HPV-unrelated
HPV-unrelated oral tongue
Germany N (%)
USN (%)
Germany N (%)
USN (%)
Germany N (%)
USN (%)
19,448 (37) 15,401 (79) 4047 (21) 1174 (6) 5296 (27) 7044 (36) 4417 (23) 1517 (8) 4397 (23) 46 (0) na na na na 7808 (40) 7188 (37) na 9 (0) 605 (6) 853 (8) 1624 (15) 7831 (72) 6467 (91) 653 (9)
21,592 (44) 17,420 (81) 4172 (19) 1241 (6) 5565 (26) 7561 (35) 4618 (21) 2607 (12) 9012 (42) 151 (1) na na na na 10,274 (48) 2148 (10) na 7 (0) 746 (6) 968 (7) 2117 (16) 9299 (71) 8546 (94) 559 (6)
25,883 (49) 20,442 (79) 5441 (21) 1613 (6) 6774 (26) 8866 (34) 6120 (24) 2510 (10) na na 2802 (11) 7852 (30) 2514 (10) 2306 (9) na na 8869 (34) 1540 (6) 1798 (15) 1332 (11) 1552 (13) 7545 (62) 5826 (88) 825 (12)
17,408 (36) 11,413 (66) 5995 (34) 541 (3) 2735 (16) 4535 (26) 4709 (27) 4888 (28) na na 2475 (14) 3989 (23) 2180 (13) 3304 (19) na na 4300 (25) 1160 (7) 1817 (23) 1244 (16) 1048 (13) 3762 (48) 3328 (91) 331 (9)
7456 (14) 5170 (69) 2286 (31) 675 (9) 1615 (22) 2270 (30) 1872 (25) 1024 (14) na 7456 (100) na na na na na na na na 944 (27) 615 (17) 598 (17) 1360 (39) 1105 (91) 108 (9)
9861 (20) 5680 (58) 4181 (42) 1172 (12) 1939 (20) 2501 (25) 2085 (21) 2164 (22) na 9861 (100) na na na na na na na na 2430 (45) 931 (17) 753 (14) 1256 (23) 1146 (94) 78 (6)
HPV, Human Papillomavirus; US, United States; na = not applicable. a For total numbers, row percentages by HPV-relation are provided for Germany and the US. b Restricted to patients diagnosed in 2004–2012, 6th and 7th version of the TNM. Patients with missing stage information (Germany: 9172 (26%), US: 4235 (14%)) were excluded. In addition, patients with ICD-O topography code C14 were excluded (Germany: 1070, US: 624), as stage was not defined for these cases in the US. For 1586 stage IV patients in Germany (10% of all stage IV patients) and 329 stage IV patients in the US (2% of all stage IV patients), information on metastasis was missing.
significantly increased between 1999 and 2013 in the US and in Germany, with an average annual increase of 3.3% (95% confidence interval (95%CI) 2.7–3.8) and 2.6% (95%CI 1.7–3.4), respectively (Fig. 1). Incidence was slightly higher in Germany throughout the whole time period. For women, incidence rates for HPV-related sites were comparable in Germany and the US in 1999–2002, but from then rose by 4.4% per year (95%CI 3.3–5.5) in Germany, while remaining stable in the US (Fig. 1). Age specific analysis revealed that this development was mostly driven by the age group 55–64, where incidence rates in Germany more than doubled during the study period (Supplementary Table 3). Divergent trends between the two countries were also observed for HPV-unrelated sites (Fig. 2). In the US, incidence rates annually declined by 2.1% (95% CI −2.8 to −1.4) for men and by 1.7% (95% CI −2.7 to −0.8) for women, while in Germany, incidence rates increased by 0.9% per year (95%CI 0.3–1.4) for men and 2.5% (95%CI 1.7–3.3) for women. Compared to the US, the incidence for HPV-unrelated sites in more recent years was more than twice as high in Germany for men, and about 50% higher for women. In HPV-unrelated oral tongue sites incidence rates slightly increased in both countries and for both sexes, albeit slightly more pronounced in Germany (Supplementary Table 4).
the SEER data. All calculations were carried out with SAS software (version 9.3), using macros developed for period analysis [32] and their model-based extensions [29]. Statistical significance was tested two-sided using a significance level of 0.05. Results Between 1997 and 2013, 19,448 (37%), 25,883 (49%) and 7456 (14%) patients with HPV-related OCPC, HPV-unrelated OCPC and HPVunrelated oral tongue cancer, respectively, were reported to the German registries. In the US, patients had relatively more often HPVrelated OCPC (21,592 (44%)) and HPV-unrelated oral tongue cancer (9861 (20%)) and less often HPV-unrelated OCPC (17,408 (36%)). Table 1 shows the characteristics of OCPC patients by HPV-relation of the cancer site in Germany and the US. For HPV-related sites, sex and stage distributions were comparable in Germany and the US. Around 80% of the patients were male and around 71% of the patients had stage IV disease at time of diagnosis. Patients in the US were slightly more often age 75+. Strong differences were observed with respect to topography with higher proportion of oropharynx cancer and lower proportion of base of tongue cancer in Germany. For HPV-unrelated OCPC, patients in Germany were more often male, were less often older than 75 years, had more often stage IV and floor of mouth and hypopharynx cancer, but less often other oral cavity cancer. For HVP-unrelated oral tongue cancer, patients in Germany were more often male, were less often age 75+, and had more often stage IV. Differences between countries were similar after stratification by sex (Supplementary Table 2).
Survival For HPV-related OCPC, age-standardized 5-year relative survival in 2010–2013 was higher in the US (60.2%) than in Germany (47.5%; Table 2). As the survival advantage was only observed in men (+16.7% units compared to −0.2% units in women), further subgroup analyses were conducted after stratification by sex. For men, survival was higher in the US in all age, topography and stage groups (all p < .0001) except in patients with distant metastasis. In adjusted models, differences were largest in men aged 15–54 years (+27.8% units) and for stage IV M0 (+26.4% units). In women, overall age-standardized 5-year relative
Incidence Age standardized incidence rates for HPV-related sites in men 10
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Fig. 1. Age-standardized incidence and average annual percentage change (AAPC) of oral cavity and pharynx cancer at HPV-related sites for male and female patients in Germany and the United States in 1999–2013.
improvements for HPV-related OCPC were comparable across age groups (+7.1 to +9.4% units) and again larger than in Germany (−9.5 to 2.8% units), leading to an increasing survival benefit for women aged 15–54 and 55–64 in the US that was, however, not significant. For elderly women, survival was significantly higher in Germany in 2002–2005 (+20.8% units) but decreased over time and was only 4.2% units higher in Germany in 2010–2013.
survival in Germany and the US were comparable (53.5 and 53.3%). Significant differences in adjusted models were only observed for stage I-III and for oropharynx cancer showing higher survival in Germany (+11.3 and +12.0% units; p = .0124 and p = .0212). Five-year age-standardized relative survival for HPV-unrelated OCPC was comparable in Germany and the US (44.0 and 45.6%; Table 2). After adjustment for stage, survival was significantly higher in Germany, overall and for males (both: +2.6% units). For HPV-unrelated oral tongue cancer, 5-year age-standardized relative survival was higher in the US than in Germany overall (+6.8% units; Table 2) and for men (+7.0% units) and women (+4.5% units). However, after adjustment for stage, survival estimates in Germany and the US were comparable with differences smaller than 1.0% unit. Country comparison of 5-year relative survival in 2002–2005 and 2006–2009 as well as survival trends between 2002–2005 and 2010–2013 are shown in Fig. 3 and Supplementary Table 5. In Germany, modest improvements in 5-year age-standardized relative survival by 2.6, 3.6 and 2.6% units were observed for HPV-related OCPC, HPV-unrelated OCPC and oral tongue cancer. In contrast, in the US, a steep rise in survival rates by 13.3% units was seen for HPV-related OCPC, whereas survival increased by about 5% units for the other sites. For HPV-related sites in men in the US, the improvements were more pronounced in older age (+8.4, +12.1 and +20.5% units for patients aged 15–54, 55–64 and 65+ years). In Germany, increases were generally smaller and no consistent age pattern was observed. Consequently, the survival difference between Germany and the US became more distinct in recent years. While in men age 15–54 years survival was already 24.3% units higher in the US in 2002–2005, men aged 65 years and older had comparable survival in 2002–2005 but much higher survival in the US in 2010–13 (+20.5% units). In US women,
Sensitivity analyses In sensitivity analyses, we restricted all analyses for the US to the white population (40.081 incident cases). The characteristics of white OCPC patients (Supplementary Table 7) were comparable with the characteristics of the total group. After restricting to whites, incidence of HPV-related cancers was higher in recent years, particularly among males, while no substantial changes were observed at HPV-unrelated sites (Supplementary Figs. 1 and 2). Survival for white OCPC patients in the US was higher compared to results for all US patients in all subgroups (Supplementary Table 7, Supplementary Fig. 3). Consequently, the previously described survival advantage for HPV-related cancers in the US compared to Germany was larger. Discussion The present study provides a comprehensive comparison of OCPC incidence and survival in Germany and the US showing strong differences across countries. For HPV-related sites in men, age-standardized incidence rates were largely comparable between both countries, with increasing trends in the US and Germany. Among women, an increasing gap between the US and Germany was observed: incidence rates 11
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Fig. 2. Age-standardized incidence and average annual percentage change (AAPC) of oral cavity and pharynx cancer at HPV-unrelated sites for male and female patients in Germany and the United States in 1999–2013.
strong geographical variations [38]. A recent international study reported substantially lower HPV-prevalence in Europe (31.1%) compared to North-America (59.3%) for the years 2002–2005/06 [16]. Therefore, it can be assumed that, compared to Germany, the proportion of HPV-positive tumors was higher in the US in our study period. It should be mentioned that these findings do not automatically indicate a higher HPV-prevalence in the US population, as the lower impact of smoking and alcohol would at least partly explain the higher relative contribution of HPV in the US. It is further known that the prevalence of oral oncogenic HPV-infection has been described to be 3-times higher among men than women. This finding is consistent with the higher risk for HPV-positive oropharyngeal cancer among men in the US [39]. Recent findings from NHANES, the U.S. National Health and Nutritional Examination Survey, indicate that the higher prevalence of oral oncogenic HPV-infection among men arises from a higher number of lifetime oral sexual partners and stronger associations between oral sex and infection among men [40,41]. HPV-associated OCPC trends might correspondingly be driven predominantly by changes in sexual behavior in men. Strong country-differences in smoking, alcohol consumption and proportion of HPV-positive tumors might also explain survival differences. Our results revealed significant differences in survival after OCPC at HPV-related sites in 2010–2013, with relative survival rates in the US being up to 27.4% units higher compared to Germany.. Strong differences were only observed in men and were largest in younger patients and stage IV M0 disease. While in men below the age of 65 years, these differences were, to a lesser extent, already present in 2002–2005, they became apparent in older men only in more recent years. It has been shown that smoking and HPV-prevalence influence the
significantly increased in Germany while remaining stable in the US. For HPV-unrelated sites, incidence was much higher in Germany than in the US, particularly in men, and the trends were divergent with increasing rates in Germany but decreasing ones in the US. For HPVunrelated oral tongue cancer, no pronounced differences were observed between the two countries. Our results on OCPC incidence are consistent with previous studies from the US [7–9] and from smaller areas in Germany [33,34]. Incidence differences between Germany and the US are likely to result from different risk factor distributions. In the US, tobacco use decreased among men between 1992/93 and 2010/11 from 26.5% to 20.1% [13,15]. In Germany, smoking prevalence for men also declined, but still remains on a higher level (1992: 36.8%, 2013: 29.0%) [14]. For women, smoking prevalence was comparable in Germany [14] and the US [13,15] in 1992 (21.8% and 21.5%). While it decreased steadily in the US (2010/11: 15.5%), it initially increased in Germany to 32% in 2003 and afterwards dropped to 20.3% in 2013. These patterns are consistent with the observed overall higher OCPC incidence in German males, especially at HPV-unrelated sites and with a comparable OCPC incidence for females between 1999 and 2002 followed by an increasing incidence in Germany in later years. In addition to differences in smoking prevalence, higher alcohol consumption in Germany is another factor that might explain higher OCPC incidence [35]. Although smoking and alcohol consumption also contribute to the disease burden of HPV-related OCPC, there is convincing evidence that HPV-infection is an independent etiologic factor for these tumors. Its importance seems to be growing, as several observations indicate that HPV-prevalence in OCPC has increased over time in different populations [33,36,37]. In a recent meta-analysis, it was estimated that 39.8% of oropharyngeal cancers could be attributed to HPV16-infection with 12
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
Table 2 Country comparisons of 5-year relative survival for male and female patients with oral cavity or pharynx cancer in Germany and the United States in 2010–2013. HPV-relation
Sex
Subgroup
Germany
Country comparison: Germany – US
US
N
RS
SE
N
RS
SE
Diffb
Adj. diffc
Adj. Pc
Overalla Overalla 15–54 years 55–64 years 65+ years Base of tonguea Tonsila Oropharynxa Stage I/II/IIIa Stage IV M0a Stage IV M1a Overalla 15–54 years 55–64 years 65+ years Base of tonguea Tonsila Oropharynxa Stage I/II/IIIa Stage IV M0a Stage IV M1a
12,603 9923 3206 3606 3111 2336 3863 3707 2147 4746 501 2680 747 888 1045 635 1207 831 660 1182 98
47.5 45.4 49.0 50.3 44.6 44.5 54.6 36.3 62.2 44.4 16.8 53.5 61.8 61.6 48.9 47.9 64.4 43.3 74.8 47.9 ∼
1.1 1.4 1.4 1.4 1.6 2.6 2.4 2.2 3.2 2.3 4.1 2.0 2.8 2.6 2.6 3.6 3.2 3.9 5.0 3.0 ∼
13,621 11,199 3339 4482 3378 4775 5327 1026 2740 6732 406 2422 644 760 1018 953 1149 302 776 1170 107
60.2 62.1 76.4 71.3 57.3 60.1 67.5 45.5 69.6 62.6 14.1 53.3 69.1 65.1 44.7 50.7 60.6 31.7 59.3 55.6 ∼
1.0 1.1 1.2 1.1 1.6 1.6 1.9 3.7 2.3 1.6 3.0 1.9 2.9 2.8 2.8 3.0 2.8 4.6 3.2 2.9 ∼
−12.7 −16.7 −27.4 −21.0 −12.7 −15.6 −12.9 −9.2 −7.4 −18.2 2.7 0.2 −7.3 −3.5 4.2 −2.8 3.8 11.6 15.5 −7.7 ∼
−17.8 −22.1 −27.8 −22.1 −11.5 −16.9 −19.0 −9.0 −12.3 −26.4 −3.2 0.8 −8.4 −1.1 5.8 −1.1 2.6 12.0 11.3 −5.4 −0.1
< .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 < .0001 .2686 .7251 .0523 .7778 .1198 .7963 .4735 .0212 .0124 .0986 .9816
Both Male Female
Overalla Overalla Overalla
16,154 12,526 3628
44.0 40.8 53.4
0.9 1.1 1.6
9049 5959 3090
45.6 42.6 51.4
1.0 1.2 1.7
−1.6 −1.8 2.0
2.6 2.6 2.8
.0106 .0365 .1972
HPV-unrelated oral tongue Both Male Female
Overalla Overalla Overalla
4696 3209 1487
56.9 53.0 64.2
1.6 2.2 2.4
5829 3347 2482
63.7 60.0 68.7
1.3 1.8 1.9
−6.8 −7.0 −4.5
−0.3 −0.1 −0.9
.8343 .9639 .7443
HPV-related Both Male
Female
HPV-unrelated
HPV, Human Papillomavirus; US, United States; N, total number of cases; RS, 5-year relative survival; SE, standard errors of the estimate of RS; Diff, Difference between RS estimates; Adj., adjusted; na, not applicable; ∼, standard error of the RS period analysis was larger than 5.0. a Age-adjusted using five age groups (15–44, 45–54, 55–64, 65–74, and 75+ years). b Differences of RS between Germany and the US based on period analysis. c Differences of RS between Germany and the US and corresponding p-value based on model based period analysis including adjustment/stratification for age and stage (significant differences in bold).
only minor, while for oral tongue tumors they diminished when adjusting for tumor stage. In contrast to survival at HPV-related sites, we could not observe different patterns in country comparisons by sex or by calendar period for HPV-unrelated OCPC supporting the hypothesis of different etiologic fractions in both countries being responsible for the observed differences in survival of HPC-related sites. Our study provides the first comparison of OCPC survival by HPVrelation of the cancer site in Germany and the US, two countries with very different smoking and HPV-prevalence. Strengths include the use of two large high quality datasets that allowed describing survival differences on the population-level and conducting detailed subgroup analyses by age, sex, topography and stage. Furthermore, we were able to investigate both incidence and survival for the same underlying population. A limitation of the study is that stage information for the US data was only available since 2004 and, thus, could not be used in trend analyses. Furthermore, a considerable number of patients had to be excluded in stage-adjusted/stratified analyses due to missing information. Another limitation is the lack of detailed treatment data, especially for Germany. Potential differences in the registration practice and coding of OCPC in Germany and the US are further limitations. The distribution of the topographies was different in Germany and the US. As we were not able to decide whether this was caused by differences in coding practice or reflects a real incidence difference, we did not include topography as adjustment factor in the models. However, higher survival rates in the US for HPV-related sites were observed for all topographies. Another issue is the use of different multiple primary rules in Germany and the
prognosis of OCPC. Smoking during therapy decreases the efficacy of radiotherapy in head and neck cancers [42]. Furthermore, smoking at diagnosis and during therapy has been reported to be associated with higher mortality after oropharyngeal cancer, independent of the tumor HPV-status [43]. Patients with HPV-positive tumors have higher response rates to chemoradiotherapy [44] and are consistently demonstrated to have a better prognosis than patients with HPV-negative tumors [43,45,46]. Thus, the differences in smoking and HPV-positive tumor prevalence might at least partly explain the observed survival benefit for HPV-related OCPC in the US compared to Germany. The finding that these differences were most pronounced for younger men supports this explanation, as HPV-positive tumors are more prevalent in this group [16,36]. Furthermore, the observation that the survival advantage in the US was already present in young men in 2002–2005 but emerged more recently for patients age 65+ is consistent with the delayed incidence increase among elderly men in the US, which was thought to be explained by a later increase in the proportion of HPVpositive tumors in this age group [9]. To explain cross-country differences in survival, differences in diagnosis and treatment, potentially reflected in clinical guidelines, have to be considered. For head and neck cancer. it has previously been noted that NCCN guidelines (US) would be more comprehensive and provide greater clarity with respect to day-to-day clinical practice than ESMO guidelines (Europe) [47]. However, survival differences caused by differences in treatment can be expected to be present in both men and women and, thus, would not explain our findings. For HPV-unrelated OCPC, survival differences across countries were 13
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
OCPC by HPV-relation of site
Five-year relative survival [%]
HPV-related, GER HPV-related, US
HPV-unrelated, GER HPV-unrelated, US
be done with caution for studies that do not consider the HPV-status. Finally, close monitoring of the epidemiology of OCPC in each country is needed and should be performed after stratification of OCPC sites according to potential HPV association.
Oral tongue, GER Oral tongue, US
80 75 70 65 60 55 50 45 40 35 30
Conflict of interest The authors declare that they have no conflict of interest. Acknowledgement 2002-2005
2006-2009
Members of the GEKID Cancer Survival Working Group: Karla Geiss, Martin Meyer (Cancer Registry of Bavaria), Andrea Eberle, Sabine Luttmann (Cancer Registry of Bremen), Roland Stabenow (Cancer Registry of Berlin and the New Federal States), Stefan Hentschel, Alice Nennecke (Hamburg Cancer Registry), Joachim Kieschke, Eunice Sirri (Cancer Registry of Lower Saxony), Bernd Holleczek (Saarland Cancer Registry), Katharina Emrich (Cancer Registry of Rhineland-Palatinate), Hiltraud Kajüter, Volkmar Mattauch (Cancer Registry of North RhineWestphalia), Alexander Katalinic, Nora Eisemann (Cancer Registry of Schleswig-Holstein), Klaus Kraywinkel (Robert Koch Institute, Berlin), Hermann Brenner, Lina Jansen and Janick Weberpals (German Cancer Research Center).
2010-2013
Calendar period OCPC at HPV-related sites in men
Five-year relative survival [%]
15-54 years, GER 15-54 years, US
55-64 years, GER 55-64 years, US
65+ years, GER 65+ years, US
2006-2009
2010-2013
80 75 70 65 60 55 50 45 40 35 30
Funding 2002-2005
This work was supported by the German Cancer Aid (Deutsche Krebshilfe) [grant number 108257 and 110446].
Calendar period
OCPC at HPV-related sites in women
Five-year relative survival [%]
15-54 years, GER 15-54 years, US
55-64 years, GER 55-64 years, US
Appendix A. Supplementary material
65+ years, GER 65+ years, US
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.oraloncology.2017.11. 015.
80 75 70 65 60 55 50 45 40 35 30
References
2002-2005
2006-2009
[1] Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, et al. GLOBOCAN 2012 v1.0, cancer incidence and mortality worldwide: IARC CancerBase No. 11 [Internet]. Lyon, France: International Agency for Research on Cancer; 2013. Available from: < http://globocan.iarc.fr > [accessed 20/07/2016]. [2] International Agency for Research on Cancer, editor. Human papillomaviruses, IARC monographs on the evaluation of carcinogenic risks to humans. Lyon, France: World Health Organization; 2007. p 670. [3] Gillison ML, Chaturvedi AK, Anderson WF, Fakhry C. Epidemiology of human papillomavirus-positive head and neck squamous cell carcinoma. J Clin Oncol 2015;33:3235–42. [4] Gillison ML, D'Souza G, Westra W, Sugar E, Xiao W, Begum S, et al. Distinct risk factor profiles for human papillomavirus type 16-positive and human papillomavirus type 16-negative head and neck cancers. J Natl Cancer Inst 2008;100:407–20. [5] Day GL, Blot WJ, Austin DF, Bernstein L, Greenberg RS, Preston-Martin S, et al. Racial differences in risk of oral and pharyngeal cancer: alcohol, tobacco, and other determinants. J Natl Cancer Inst 1993;85:465–73. [6] Mehanna H, Beech T, Nicholson T, El-Hariry I, McConkey C, Paleri V, et al. Prevalence of human papillomavirus in oropharyngeal and nonoropharyngeal head and neck cancer – systematic review and meta-analysis of trends by time and region. Head Neck 2013;35:747–55. [7] Chaturvedi AK, Anderson WF, Lortet-Tieulent J, Curado MP, Ferlay J, Franceschi S, et al. Worldwide trends in incidence rates for oral cavity and oropharyngeal cancers. J Clin Oncol 2013;31:4550–9. [8] Chaturvedi AK, Engels EA, Anderson WF, Gillison ML. Incidence trends for human papillomavirus-related and -unrelated oral squamous cell carcinomas in the United States. J Clin Oncol 2008;26:612–9. [9] Zumsteg ZS, Cook-Wiens G, Yoshida E, Shiao SL, Lee NY, Mita A, et al. Incidence of oropharyngeal cancer among elderly patients in the United States. JAMA Oncol 2016;2:1617–23. [10] Gatta G, Botta L, Sanchez MJ, Anderson LA, Pierannunzio D, Licitra L, et al. Prognoses and improvement for head and neck cancers diagnosed in Europe in early 2000s: the EUROCARE-5 population-based study. Eur J Cancer 2015. [11] O'Rorke MA, Ellison MV, Murray LJ, Moran M, James J, Anderson LA. Human papillomavirus related head and neck cancer survival: a systematic review and meta-analysis. Oral Oncol 2012;48:1191–201. [12] Gillison ML, Restighini C. Anticipation of the impact of human papillomavirus on clinical decision making for the head and neck cancer patient. Hematol Oncol Clin
2010-2013
Calendar period Fig. 3. Age-standardized overall relative survival in 2002–2005, 2006–2009 and 2010–2013 in Germany and the United States for oral cavity and pharynx cancer at HPVrelated sites, HPV-unrelated sites and at HPV-unrelated oral tongue and for oral cavity and pharynx cancer at HPV-related sites by sex and age.
US. However, it has been shown that the difference between these two rules has only minor effects for OCPCs [48]. Differences in the completeness of registration might also bias the comparison. In both countries, the proportion of DCO cases cannot reliably be incorporated in our study, as cases were selected by ICD-O-3 histology codes that are very rarely specified on death certificates. However, in previous studies it has been shown that DCO proportions are generally higher in Germany than in the US [18], potentially leading to an overestimation of survival and underestimation of incidence in Germany. Thus, the reported survival advantage for HPV-related OCPC in the US might even be larger. Conclusion Due to the observed trends that indicate differences in patient characteristics with respect to HPV prevalence across countries, country comparisons should be stratified by HPV status or at least by HPV-associations of the cancer site. Furthermore, translation of evidence, e.g. with respect to prognosis and treatment, across countries should only 14
Oral Oncology 76 (2018) 8–15
L. Jansen et al.
[31] Sobin LH, Gospodarowicz MK, Wittekind C. TNM classification of malignant tumors, 7th ed. Oxford, UK: Wiley-Blackwell; 2009. [32] Brenner H, Gefeller O, Hakulinen T. A computer program for period analysis of cancer patient survival. Eur J Cancer 2002;38:690–5. [33] Tinhofer I, Johrens K, Keilholz U, Kaufmann A, Lehmann A, Weichert W, et al. Contribution of human papilloma virus to the incidence of squamous cell carcinoma of the head and neck in a European population with high smoking prevalence. Eur J Cancer 2015;51:514–21. [34] Bayer O, Kruger M, Koutsimpelas D, Emrich K, Ressing M, Zeissig SR, et al. Changes in incidence and mortality trends of head and neck cancer in Rhineland-palatinate, 2000–2009. Laryngorhinootologie 2015;94:451–8. [35] World Health Organization. Global status report on alcohol and health 2014. World Health Organization; 2014. [36] Castellsague X, Alemany L, Quer M, Halec G, Quiros B, Tous S, et al. HPV involvement in head and neck cancers: comprehensive assessment of biomarkers in 3680 patients. J Natl Cancer Inst 2016;108. djv403. [37] Chaturvedi AK, Engels EA, Pfeiffer RM, Hernandez BY, Xiao W, Kim E, et al. Human papillomavirus and rising oropharyngeal cancer incidence in the United States. J Clin Oncol 2011;29:4294–301. [38] Ndiaye C, Mena M, Alemany L, Arbyn M, Castellsague X, Laporte L, et al. HPV DNA, E6/E7 mRNA, and p16INK4a detection in head and neck cancers: a systematic review and meta-analysis. Lancet Oncol 2014;15:1319–31. [39] Gillison ML, Broutian T, Pickard RK, Tong ZY, Xiao W, Kahle L, et al. Prevalence of oral HPV infection in the United States, 2009–2010. JAMA 2012;307:693–703. [40] D'Souza G, Cullen K, Bowie J, Thorpe R, Fakhry C. Differences in oral sexual behaviors by gender, age, and race explain observed differences in prevalence of oral human papillomavirus infection. PLoS ONE 2014;9:e86023. [41] Chaturvedi AK, Graubard BI, Broutian T, Pickard RK, Tong ZY, Xiao W, et al. NHANES 2009–2012 findings: association of sexual behaviors with higher prevalence of oral oncogenic human papillomavirus infections in U.S. men. Can Res 2015;75:2468–77. [42] Browman GP, Wong G, Hodson I, Sathya J, Russell R, McAlpine L, et al. Influence of cigarette smoking on the efficacy of radiation therapy in head and neck cancer. N Engl J Med 1993;328:159–63. [43] Gillison ML, Zhang Q, Jordan R, Xiao W, Westra WH, Trotti A, et al. Tobacco smoking and increased risk of death and progression for patients with p16-positive and p16-negative oropharyngeal cancer. J Clin Oncol 2012;30:2102–11. [44] Fakhry C, Westra WH, Li S, Cmelak A, Ridge JA, Pinto H, et al. Improved survival of patients with human papillomavirus-positive head and neck squamous cell carcinoma in a prospective clinical trial. J Natl Cancer Inst 2008;100:261–9. [45] Goodman MT, Saraiya M, Thompson TD, Steinau M, Hernandez BY, Lynch CF, et al. Human papillomavirus genotype and oropharynx cancer survival in the United States of America. Eur J Cancer 2015;51:2759–67. [46] D'Souza G, Anantharaman D, Gheit T, Abedi-Ardekani B, Beachler DC, Conway DI, et al. Effect of HPV on head and neck cancer patient survival, by region and tumor site: a comparison of 1362 cases across three continents. Oral Oncol 2016;62:20–7. [47] D'Cruz A, Lin T, Anand AK, Atmakusuma D, Calaguas MJ, Chitapanarux I, et al. Consensus recommendations for management of head and neck cancer in Asian countries: a review of international guidelines. Oral Oncol 2013;49:872–7. [48] Weir HK, Johnson CJ, Thompson TD. The effect of multiple primary rules on population-based cancer survival. Cancer Causes Control 2013;24:1231–42.
North Am 2015;29:1045–60. [13] US Department of Commerce, Census Bureau. National Cancer Institute-sponsored Tobacco Use Supplement to the Current Population Survey (2010–11); 2012. < http://appliedresearch.cancer.gov/tus-cps/results/data1011/table 1. html > [accessed 28/07/2016]. [14] Microcensus – health questions. Federal Statistical Office, Branch Bonn [accessed 06/04/2017]. [15] Jemal A, Thun M, Yu XQ, Hartman AM, Cokkinides V, Center MM, et al. Changes in smoking prevalence among U.S. adults by state and region: estimates from the Tobacco Use Supplement to the Current Population Survey, 1992–2007. BMC Public Health 2011;11:512. [16] Anantharaman D, Abedi-Ardekani B, Beachler DC, Gheit T, Olshan AF, Wisniewski K, et al. Geographic heterogeneity in the prevalence of human papillomavirus in head and neck cancer. Int J Cancer 2017;140:1968–75. [17] Hiripi E, Gondos A, Emrich K, Holleczek B, Katalinic A, Luttmann S, et al. Survival from common and rare cancers in Germany in the early 21st century. Ann Oncol 2011;23:472–9. [18] Jansen L, Castro FA, Gondos A, Krilaviciute A, Barnes B, Eberle A, et al. Recent cancer survival in Germany: an analysis of common and less common cancers. Int J Cancer 2015;136:2649–58. [19] Brown LM, Check DP, Devesa SS. Oral cavity and pharynx cancer incidence trends by subsite in the United States: changing gender patterns. J Oncol 2012;2012:649498. [20] Fritz APC, Jack A, Shanmugaratnam K, Sobin LH, Parkin DM, et al. International classification of diseases for oncology. 3rd ed. Geneva: World Health Organization; 2000. [21] Surveillance, Epidemiology, and End Results (SEER) Program Research Data (1973–2013). National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2016, based on the November 2015 submission. < www.seer.cancer.gov > . [22] Bray F, Ferlay J. Age-standardisation. In: Forman DFB, Brewster DH, Gombe Mbalawa C, Kohler B, Piñeros M, Steliarova-Foucher E, editors. Cancer incidence in five continents Vol X. Lyon: IARC; 2007 [chapter 7]. [23] Statistical research and applications branch NCI. Joinpoint regression program. 4.0. 4. ed2013. [24] OFFIS e.V. CARESS. Analytisches Informationssystem für Krebsregisterdaten. 8.8. 12.6 ed. Oldenburg, Germany; 2016. [25] Brenner H, Gefeller O, Hakulinen T. Period analysis for ‘up-to-date' cancer survival data: theory, empirical evaluation, computational realisation and applications. Eur J Cancer 2004;40:326–35. [26] Ederer F, Heise H. Instructions to IBM 650 programmers in processing survival computations. Methodological note no. 10. End results evaluation section. Bethesda, MD: National Cancer Institute; 1959. [27] Center for Disease Control and Prevention Publications and Information Products Life Tables. < http://www.cdc.gov/nchs/products/life_tables.htm > [accessed 01/ 07/2016]. [28] Corazziari I, Quinn M, Capocaccia R. Standard cancer patient population for age standardising survival ratios. Eur J Cancer 2004;40:2307–16. [29] Brenner H, Hakulinen T. Up-to-date and precise estimates of cancer patient survival: model-based period analysis. Am J Epidemiol 2006;164:689–96. [30] Sobin LH, Wittekind C. TNM classification of malignant tumours. 6th ed. New York: Wiley-Liss; 2002.
15