Long-term trends in pancreatic cancer mortality in Spain (1952–2012)

Long-term trends in pancreatic cancer mortality in Spain (1952–2012) Background: Pancreatic cancer is acquiring increasing prominence as a cause of cancer death in the population. The purpose of this study was to analyze long-term pancreatic cancer mortality trends in Spain and evaluate the independent effects of age, death period and birth cohort on these trends. Methods: Population and mortality data for the period 1952–2012 were obtained from the Spanish National Statistics Institute. Pancreatic cancer deaths were identified using the International Classification of Diseases ICD-6 to ICD-9 (157 code) and ICD-10 (C25 code). Age-specific and age-adjusted mortality rates were computed by sex, region and five-year period. Changes in pancreatic cancer mortality trends were evaluated using joinpoint regression analyses by sex and region. Age-period-cohort log-linear models were fitted separately for each sex, and segmented regression models were used to detect changes in period- and cohort-effect curvatures. Results: In men, rates increased by 4.1% per annum from 1975 until the mid-1980s and by 1.1% thereafter. In women, there was an increase of 3.6% per annum until the late 1980s, and 1.4% per annum from 1987 to 2012. With reference to the cohort effects, there was an increase in mortality until the generations born in the 1950s in men and a subsequent decline detected by the change point in 1960. A similar trend was observed in women, but the change point occurred 10 years later than in men. Conclusions: Pancreatic cancer mortality increased over the study period in both sexes and all regions. An important rise in rates -around 4% annually- was registered until the 1980s, and upward trends were more moderate subsequently. The differences among sexes in trends in younger generations may be linked to different past prevalence of exposure to some risk factors, particularly tobacco, which underwentan earlier decrease in menthaninwomen. Keywords: Pancreatic cancer, Tobacco smoking, Mortality, Age-period-cohort analysis, Change-points, Time trends, Spain Background males and 7.6 cases per 100,000 females [1]. According to Though, in terms of incidence, pancreatic cancer is not these data, this cancer has become the tenth most com- among the most frequent cancers, its high lethality mon cancer type registered among men and the sixth places this malignant tumor among those that cause a among women. higher number of deaths worldwide [1]. The overall As regards to mortality, in 2012 pancreatic cancer prognosis of pancreatic cancer is extremely poor, with ranked seventh as cause of cancer death among Spanish five-year relative survival rates around 6% in Europe [2]. men, with age adjusted mortality rates of 10.7 per 100,000 In Spain, 6367 new pancreatic cancer cases were esti- inhabitants [3], being the third leading cause of oncologic mated to occur in 2012, with age adjusted incidence rates deaths among men between 40 and 59 years [4]. In (European standard population) of 11.5 cases per 100,000 women, pancreatic cancer was the fourth most common cause of oncologic death, with rates around 6.8 per 100,000. In sum, pancreatic cancer is currently responsible * Correspondence: nuria.aragones@salud.madrid.org Cancer and Environmental Epidemiology Unit, National Center for for 5 and 7% of the total number of deaths due to cancer Epidemiology, Carlos III Institute of Health, Madrid, Spain among Spanish males and females, respectively. Consortium for Biomedical Research in Epidemiology and Public Health (CIBER Epidemiología y Salud Pública, CIBERESP), Madrid, Spain Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 2 of 10 The etiology of pancreatic cancer is unclear [5]and, as time trends for each region and sex. We used the joinpoint in other cancers, probably multifactorial. Several factors regression analysis to evaluate the presence of change have been suggested as possible causes for this neoplasm, points in adjusted mortality rates over time by sex and re- but their contribution, according to their relative risks, is gion and to estimate the annual percent of change over the small [6]. Between 3 and 7% of cases could be associated study period [11]. Ceuta and Melilla regions were excluded with genetic susceptibility. Regarding exogenous expo- from this analysis, because of their small populations. sures, tobacco smoking (with strong evidence) and Helico- With respect to age-period-cohort models, first, bacter pylori infection (with moderate evidence) have age-specific mortality rates per 100,000 person-years were been considered the major risk factors for pancreatic can- computed by sex and calendar period (using five-year pe- cer, in terms of their population attributable fraction, in a riods) at the national level. Then, separate log-linear Pois- recent review [6]. According to the American Institute for son models were fitted to study the effect of age, period of Cancer Research, there is also convincing evidence to con- death and birth cohort for each sex on mortality trends. sider body fatness as a risk factor for pancreatic cancer To address the “non-identifiability” problem (i.e. the three [7]. Other exposures or clinical entities that have been factors -age, period and cohort- are linearly dependent), suggested to be associated with increased risk include high we used Osmond and Gardner’ssolution[12], as well as red meat consumption, type II Diabetes Mellitus, chronic curvature effects and net drift as proposed by Holford pancreatitis, high alcohol consumption, hepatitis B virus [13]. The Osmond-Gardner solution splits net drift into infection and specific occupational exposures, such as cohort and period slopes, by minimizing any disagreement certain pesticides, organic solvents, polycyclic aromatic in parameter estimates between the full three-factor hydrocarbons and nickel compounds [6–9]. On the other model and each of the two-factor models (age-period, hand, several factors could have a preventive effect, as it is age-cohort and period-cohort) according to their good- the case of allergic conditions, high fruit and vegetable ness of fit. Moreover, it allows to estimate two parameters consumption and physical activity [6–8]. not affected by the non-identifiability problem: (i) overall This study aimed to monitor pancreatic cancer mortal- change over time (denominated net drift), which is the ity trends since the middle of the twentieth century until sum of the cohort and period slopes [13]; and (ii) devi- recent years in Spain, using joinpoint regression models ation of any period or cohort estimates from the general and age-period-cohort analyses to evaluate the inde- trend (denominated curvature). Age groups < 30 years pendent effects of age, death period and birth cohort on were excluded from this analysis due to the limited num- these trends. ber of deaths. The open-ended category of persons aged 85 years and over was also excluded. We checked for Methods extra-Poisson dispersion [14] and, where present, effects Mortality data for the calendar period 1952–2012 were were calculated using a negative binomial distribution. obtained from the Spanish National Statistics Institute The presence of change points and 95% confidence inter- (Instituto Nacional de Estadística) at national level. Dur- vals in the curvatures of the cohort and period effects was ing this period, different revisions of the International evaluated by fitting segmented models to the relationship Classification of Diseases (ICD) have been used. Codes between curvature effects and time. Details of the recursive selected to identify deaths due to pancreatic cancer were algorithm used to estimate the segmented regression have adapted accordingly: code 157 in the ICD-6 to ICD-9 been published elsewhere [15], and the procedure can be and code C25 in the ICD-10. Population data corre- easily fitted using the R package “segmented” [16]. sponding to censuses and municipal rolls for the mid- year of each quinquennium were also obtained from the Results Spanish National Statistics Institute. As a geographic reference, Fig. 1 shows the location of the From 1975 to 2012, mortality and population data are Spanish Autonomous Communities, with the regional dis- public and available at regional level, and were stratified by tribution of pancreatic cancer mortality in both sexes in sex, five-year-age group (from 0 to 4 to 85+ years), calendar the last quinquennium (2008–2012). AAMRs (1976 ESP) year and region (Autonomous Community). Age-adjusted by sex, Autonomous Community and calendar period are mortality rates (AAMR) per 100,000 person-years were presented in Table 1. AAMRs have increased in both sexes then calculated, by the direct method, for each sex, in all regions, but not uniformly. The largest increases oc- five-year calendar period and region, using the 1976 Euro- curred before the 1990s in both sexes. Then, pancreatic pean Standard Population (ESP). Age-adjusted mortality cancer mortality rates have increased at a slower speed. rates were also calculated using the 2013 ESP to allow com- No differences in this trend were observed between men parison with other works (Additional files 1 and 2)[10]. and women. In both sexes, the highest increment took Additionally, annual age-adjusted mortality rates and their place between the 1978–1982 and the 1983–1987 quin- corresponding standard errors were calculated to study quennia (18% in men and 23% in women), with more Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 3 of 10 Asturias Cantabria Basque Country Galicia Navarre La Rioja Catalonia Castile and Leon Aragon Madrid Castile Balearic Islands Valencian La Mancha Community Extremadura Murcia Andalousia [6.11,7.78] (7.78,8.53] (8.53,8.65] Canary Islands Ceuta/Melilla (8.65,8.97] (8.97,9.71] Fig. 1 Pancreatic cancer mortality in Spain (2008–2012): AAMR per 100,000 person-years (1976 ESP) by Autonomous Community modest increments thereafter (between 5 and 11%). Differ- late 1980s, and then increased by 1.4% per annum from ences among regions over the study period have been 1987 to the end of the study period. By region, some of slightly reduced: while in the quinquennium 1978–1982 them also showed a two-phase pattern, although with dif- the ratio between the highest and lowest rates was around ferences in the year when the change was estimated to 1.8 in males and 1.7 in females, in 2008–2012 this ratio was have occurred. Meanwhile, in others no inflection points around 1.3 in both sexes. In men, the highest rates in the were detected and rates increased at the same speed quinquennium 2008–2012 were found in Asturias, La Rioja through the study period. Asturias was the Autonomous and Galicia, whereas in women the highest rates Community with the smaller overall increase (the only corresponded to Navarra, Asturias and Cantabria. Since a one under 1%) in both sexes, though their mortality rates new European Standard Population has been published re- were among the highest of the country in all quinquennia. cently, AAMRs were recalculated using the 2013 ESP Figure 2 shows the evolution of smoothed pancreatic can- (Additional file 1: Figure S1 and Additional file 2:Table S1). cer death rates over time by sex and region. The presence Results are similar, though AAMRs tend to be higher when of changes in trends is visible. using the 2013 ESP, given the greater weight that this new Figure 3 depicts age-specific rates by birth cohort, in standard population gives to older age groups. men and women. In both sexes, clear increases over time The results from joinpoint regression analyses over the are observed in all age groups over 45 years of age. How- period 1975–2012 by Autonomous Community and sex, ever, in males younger than 45 years of age a trend to and in Spain as a whole are presented in Table 2.Bothin stabilization, or even a reduction, is observed, while men and women statistically significant upward trends among women this trend to stabilization is only observed were seen in nearly all Autonomous Communities. In for the 30–35 age group. Nevertheless, trends in the youn- men, there was an overall increase in the rates of 2% per gest age groups are based on a small number of deaths. annum. Joinpoint analysis detected a change point in the Table 3 presents the goodness of fit of the different mid-1980s: during the first period, rates increased by 4.1% age-period-cohort models. In men, period effect was the per annum and by 1.1% during the second period. In main contributor to the estimated trend in mortality, while women, pancreatic cancer mortality rates experienced a in women the main contributor was the cohort effect. marked increase of 3.6% per annum from 1975 until the Figure 4 depicts cohort and period effects with the change Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 4 of 10 Table 1 Pancreatic cancer mortality in Spain (1978–2012) by sex, Autonomous Community and calendar period 1978–82 1983–87 1988–92 1993–97 1998–02 2003–07 2008–12 Men Andalucía 5.34 6.18 6.71 7.37 7.57 8.29 9.34 Aragón 5.91 7.57 9.39 8.95 9.83 10.38 11.17 Asturias 8.81 11.67 10.26 9.79 11.10 11.64 12.37 C.Valenciana 6.33 7.84 7.71 8.45 9.33 9.04 10.60 Cantabria 7.88 9.49 10.78 10.74 10.33 11.79 10.81 Castilla y León 6.43 7.36 8.39 9.38 9.06 10.46 10.59 Castilla-la Mancha 4.86 5.96 6.78 7.18 7.74 8.68 9.26 Cataluña 6.91 8.34 8.95 9.67 9.44 10.07 10.79 Ceuta 5.14 10.31 11.69 14.56 13.21 13.66 9.54 Extremadura 6.18 6.86 9.18 8.86 10.50 9.80 11.36 Galicia 5.90 8.14 8.54 9.70 10.85 10.99 11.47 Islas Baleares 6.57 7.89 9.17 7.99 9.38 9.81 9.57 Islas Canarias 9.26 9.30 10.77 9.66 10.38 10.55 10.21 La Rioja 7.77 8.01 10.43 9.67 9.08 10.46 12.17 Madrid 5.93 5.76 7.99 7.92 9.31 9.52 9.19 Melilla 1.68 2.09 0.87 3.77 8.49 8.34 8.39 Murcia 6.16 6.33 6.48 7.73 9.06 9.66 10.21 Navarra 6.49 8.25 9.89 10.98 11.37 10.87 11.30 País Vasco 7.28 8.89 9.49 10.74 10.42 10.65 10.75 Spain 6.33 7.49 8.31 8.81 9.32 9.77 10.32 Women Andalucía 3.41 4.11 4.37 4.75 4.72 5.02 5.97 Aragón 3.83 4.85 5.14 6.01 5.87 6.33 6.37 Asturias 5.27 6.51 5.80 6.02 6.03 6.71 7.26 C.Valenciana 3.17 4.37 4.75 5.34 5.45 6.03 6.69 Cantabria 4.34 5.44 6.52 8.17 5.97 7.19 7.14 Castilla y León 3.57 4.61 4.74 5.35 5.69 5.77 6.81 Castilla-la Mancha 3.12 3.75 4.87 4.69 4.89 5.75 6.53 Cataluña 3.97 4.80 5.21 5.28 5.88 6.12 6.69 Ceuta 6.39 7.07 7.90 6.99 8.08 6.28 4.56 Extremadura 4.18 4.12 4.92 5.55 5.39 6.15 6.88 Galicia 3.35 4.36 4.50 5.47 5.89 6.16 6.58 Islas Baleares 3.52 4.44 4.95 5.77 5.58 5.98 6.01 Islas Canarias 3.74 5.06 6.24 6.42 6.58 6.91 6.69 La Rioja 4.46 5.57 5.53 5.44 6.95 6.27 6.61 Madrid 3.19 3.48 4.49 5.09 5.61 5.84 6.47 Melilla 3.45 1.25 3.63 5.18 4.28 2.97 2.46 Murcia 4.51 3.59 4.03 5.08 4.99 5.46 6.64 Navarra 3.86 5.28 5.12 6.49 6.69 6.97 8.18 País Vasco 3.86 5.25 5.74 5.81 6.47 6.44 6.80 Spain 3.64 4.46 4.89 5.36 5.61 5.96 6.57 Age-adjusted mortality rates per 100,000 person-years (1976 ESP) Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 5 of 10 Table 2 Pancreatic cancer mortality trend changes in Spain evaluated using joinpoint analysis by sex and Autonomous Community Period 1 Period 2 APC N. of change points Years APC 1 Years APC 2 a a Men Andalucía 2.89 1 1975-1978 17.66 1978-2012 1.68 Aragón 1.71 0- - - - Asturias 0.84 0- - - - a a a C. Valenciana 2.14 1 1975-1985 4.77 1985-2012 1.19 Cantabria 0.84 0- - - - a a a Castilla y León 1.93 1 1975-1990 3.39 1990-2012 0.95 Castilla-la Mancha 2.00 0- - - - a a a Cataluña 1.79 1 1975-1987 3.90 1987-2012 0.80 Extremadura 1.80 0- - - - a a a Galicia 2.82 1 1975-1985 6.93 1985-2012 1.34 a a Islas Baleares 2.21 1 1975-1986 6.51 1986-2012 0.45 Islas Canarias 2.37 1 1975-1978 29.02 1978-2012 0.30 La Rioja 1.38 0- - - - a a Madrid 1.62 1 1975-2001 2.50 2001-2012 -0.42 Murcia 1.74 0- - - - Navarra 1.39 0- - - - País Vasco 1.03 0- - - - a a a Spain 1.99 1 1975-1986 4.10 1986-2012 1.11 Women Andalucía 1.57 0- - - - Aragón 1.71 0- - - - Asturias 0.94 0- - - - a a a C. Valenciana 2.78 1 1975-1984 6.48 1984-2012 1.61 a a Cantabria 2.23 1 1975-1991 5.55 1991-2012 -0.23 Castilla y León 1.89 0- - - - Castilla-la Mancha 2.34 0- - - - Cataluña 1.52 0- - - - Extremadura 1.65 0- - - - a a Galicia 3.70 1 1975-1977 38.72 1977-2012 1.99 Islas Baleares 1.54 0- - - - a a Islas Canarias 1.76 1 1975-1992 3.64 1992-2012 0.18 La Rioja 0.84 0 - - - - a a a Madrid 2.55 1 1975-1994 3.61 1994-2012 1.43 Murcia 1.65 0- - - - Navarra 2.08 0- - - - a a a País Vasco 1.66 1 1975-1989 3.03 1989-2012 0.84 a a a Spain 2.09 1 1975-1987 3.56 1987-2012 1.39 APC Annual Percentage of Change Statistically significant trend as obtained from the segmented regression. Statistical tests were two sided. The significance level was considered as 0.05 points (listed in Table 4) detected on their curvatures. With 1960. In women, as can be seen from its flatter curvature respect to the cohort effect in men, there was an increase (thin line), the cohort effect is less pronounced and the in mortality until the generations born in the 1950s and a change point was placed a decade later, though this subsequent decline with a detectable change point around finding is difficult to assess as it is based on very few Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 6 of 10 Men Women La Rioja Aragon Galicia C.Valenciana Extremadura Castilla y León Murcia Cataluña País Vasco Andalucia Navarra Castilla−la Mancha Asturias Madrid Islas Baleares Cantabria Islas Canarias Navarra Castilla y León Galicia Cantabria Castilla−la Mancha C.Valenciana La Rioja Extremadura Madrid Andalucia Murcia Cataluña Aragon País Vasco Islas Canarias Asturias Islas Baleares 1975 1985 1995 2005 2015 1975 1985 1995 2005 2015 Calendar Period Fig. 2 Pancreatic cancer mortality trends in Spain (1975–2012), by sex and Autonomous Community. Age-adjusted smoothed mortality rates per 100,000 person-years (1976 ESP) Pancreas men Pancreas women 85+ 80− 85+ 75− 80− 70− 75− 65− 70− 60− 65− 55− 60− 50− 55− 50− 45− 45− 40− 40− 35− 35− 30− 30− 25− 1860 1880 1900 1920 1940 1960 1980 1860 1880 1900 1920 1940 1960 1980 Year of birth Year of birth Fig. 3 AAMR per 100,000 person-years for pancreatic cancer by birth cohort and sex, Spain (1952–2012) Age−specific rates x 100,000 Age standardized death rates (per 100,000 person−years) 1e−02 1e+00 1e+02 Age−specific rates x 100,000 1e−02 1e+00 1e+02 Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 7 of 10 Table 3 Goodness of fit in age, period and cohort models for pancreatic cancer mortality by sex, Spain (1952–2012) Degrees of freedom Deviance % of change in deviance Men Age 121 9980 – Age + drift 120 1382 Reference Age + period 110 354 74.4% Age + cohort 100 509 63.2% Age + period + cohort 90 109 92.1% Women Age 121 6094 – Age + drift 120 694 Reference Age + period 110 327 52.9% Age + cohort 100 153 78.0% Age + period + cohort 90 82 88.2% Pancreas men period cohort 1880 1900 1920 1940 1960 1980 2000 2020 Y Year ear Pancreas women period cohort 1880 1900 1920 1940 1960 1980 2000 2020 Y Year ear Fig. 4 Cohort and period effects on pancreatic cancer mortality by sex, Spain (1952–2012). Cohort and period effects (thick lines), curvature (thin lines) and change points in the curvatures (vertical grey lines) 0.2 0.5 1.0 2.0 4.0 0.2 0.5 1.0 2.0 4.0 Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 8 of 10 Table 4 Cohort and period effect curvature change-points on with the different evolution in the period effect between pancreatic cancer mortality by sex, Spain (1952–2012) men and women in the last years, with a lower annual in- Changes in cohort effect crease of rates in men than in women since the 1990s. In Europe, there are geographical differences in pancre- Birth year Birth year (95% CI) (95% CI) atic cancer mortality trends. Mortality rates have increased in the last three decades in countries like France, Germany, Men 1904.6 (1898.7–1910.5) 1960.6 (1958.9–1962.3) Greece, Italy, Romania or Bulgaria, whereas in Sweden, the Women 1894.4 (1890.9–1897.8) 1971.2 (1968.9–1973.6) United Kingdom and Norway they have decreased (in the last two countries only in men). In Denmark, Ireland, Changes in period effect Finland and Holland, mortality in men diminished until Year of death Year of death 1990s and then began to increase [20]. In the United States, (95% CI) (95% CI) white and black people show opposite trends: mortality Men 1963.5 (1961.7–1965.2) 1987.83 (1985.52–1990.14) rates in white men decreased from 1970 to 1995, and have Women 1964.0 (1957.1–1970.9) increased since then; in white women, there was a slight in- Year of birth with significant trend change as obtained from the segmented crease between 1970 and 1984, a stabilization until the late regression analysis of cohort curvatures from the three-factor model Year of death with significant trend change as obtained from the segmented 1990s, and an increase thereafter. On the other hand, in regression analysis of period curvatures from the three-factor model black men and women, rates increased until the late 1980s – early 1990s and have decreased since then [21]. In rates and deaths. For the period effect, there is a Canada, mortality rates between 1992 and 2009 declined in change in the general trend around 1962–1965 in men and remained stable in women [22]. both sexes, and a second change point in men around Understanding the reasons of the increase in pancre- 1988, not visible in women. atic cancer mortality is challenging, given the complex and not well understood etiology of this neoplasm. In Discussion contrast with other cancer locations, like lung or cervical Our results show that pancreatic cancer mortality rates cancer, which are mainly associated with a unique risk increased over the study period in Spain in both sexes factor, pancreatic cancer has been associated with mul- and all regions. In general, this rise was similar in both tiple factors with modest effect sizes and, some of them, sexes and in areas with higher and lower rates compared with high prevalence of exposure in the general popula- to Spain as a whole. In men, age-adjusted mortality rates tion [6]. Probably, differences in latency periods among annually grew on average by 4.1% in the period 1975– these factors may also have a different effect on pancre- 86, and by 1.1% between 1986 and 2012. In women there atic cancer trends, obscuring their specific contribution. was a similar trend, with a bigger increase until the late Among the risk factors for this cancer, the only univer- 1980s (3.6% annually between 1975 and 1987) than in sally accepted one is tobacco consumption [6]. Trends of the years after (1.4% in the period 1987–2012). pancreatic cancer mortality do not resemble those of other The age-period-cohort analysis shows that, taking the tumors strongly related to smoking, such as lung cancer, mean rate for all cohorts as reference, the risk of dying that is decreasing in males and increasing in females in from pancreatic cancer increased in generations born be- Spain and in most developed countries. Nevertheless, the tween 1870 and 1960 in men and women. This similarity slower increase in pancreatic cancer mortality rates in the in the trends could be related to changes in the exposure last decades could be influenced by the decreasing preva- to risk factors linked to birth cohort and shared by both lence of tobacco consumption. In Spain, prevalence of sexes. The rates observed in young men may indicate a smoking shows a decreasing trend in men since the late levelling off in mortality in the most recent generations. eighties (not previous data available), while in women, Among women, this phenomenon is less clear, though the prevalence rose until the late nineties and then slightly de- stabilization of the rates in the youngest age group could creased [23]. Accordingly, the continuous decrease of the indicate that trend might soon parallel that of men. rates in men cohorts born since the 1960s -not so evident For the period effect, there is a change in the general in women-, could be related to the different evolution in trend around 1962–1965 in males and females and a sec- the prevalence of smoking between sexes. Considering ond change point in men around 1988. Though survival 2013 data, smoking prevalence in men over 34 years old in pancreatic cancer is still very low, a slight improvement had fell 9.8 percentage points in the last decade, while in has been described in some countries in the last years women over 34 years old it had not decreased [24]. This [17–19], which could have contributed to the slower in- would be consistent with a role of tobacco exposure in crease in the mortality rates since the 1990s observed in pancreatic cancer risk, but with the existence of other im- our study. The improvement in survival has been reported portant contributing risk factors that would counterbal- to be slightly higher in men [17]. This would be consistent ance the effect of the reduction in tobacco exposure. Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 9 of 10 Another suspected risk factor for pancreatic cancer is Conclusions Helicobacter pylori infection [6]. Again, trends of pan- This study summarizes the trends in pancreatic cancer creatic cancer mortality do not resemble those of gastric mortality in Spain, allowing for a detailed analysis of the cancer, strongly associated with this infection and whose influence of age, period and cohort effects in each sex. mortality rates are decreasing in both sexes. Neverthe- Like in other developed countries, pancreatic cancer mor- less, according to Fig. 4, the higher increase in the risk tality has been increasing over the last decades. However, in cohorts born until the late nineteenth century-early differences have been identified between sexes. In men, twentieth century and the decrease in those born since mortality rates show a stabilization or even a decrease the 1960s or 1970s in men and women, respectively, is since the early 1990s, mainly among post-1960 birth co- congruent with the epidemiology of H. pylori infection. horts, while in women, a stabilization in the trend is ob- Though this effect is not as clear as it is in gastric cancer servable only among the youngest generations (born after [25], which could be explained by a weaker association the late 1970s). These differences may partially mirror the between H. pylori infection and pancreatic cancer, there evolution of some established risk factors for pancreatic seems to be some coincidence in time between both tu- cancer, such as tobacco exposure, in men and women. mors, especially in women. Further research on the causes of pancreatic cancer is Though with limited evidence, other factors have been needed. In the meanwhile, recommendations to reduce suggested to play a role in the etiology of pancreatic can- exposure to preventable risk factors that have been associ- cer. Among proposed risk factors, extensive research has ated with pancreatic cancer, such as tobacco smoking, focused on the role of diet and anthropometric factors obesity, diet and lack of physical activity may serve to re- [26, 27]. As possible protective factors, a medical history duce the impact of this and other chronic and malignant of allergy, the consumption of fruits and vegetables, diseases. physical activity and parity have been suggested [6, 28– 31]. The controversial evidence from epidemiologic Additional files studies, the diverging trends of this broad spectrum of factors and the different prevalence of exposure among Additional file 1 Figure S1. Pancreatic cancer mortality in Spain (2008–2012): AAMR per 100,000 person-years (2013 ESP) by Autonomous regions make it not easy to disentangle the specific role Community. (PDF 529 kb) of each factor in pancreatic cancer trends. Additional file 2 Table S1. Pancreatic cancer mortality in Spain. AAMR A different aspect that could have played a role in the per 100,000 person-years (2013 ESP) by sex, Autonomous Community observed upward trend in pancreatic cancer mortality is and calendar period. (DOCX 30 kb) the introduction of Computerized Tomography in Spain during the late 1970s of the twentieth century, and the Abbreviations AAMR: Age-adjusted mortality rates; ESP: European Standard Population; spreading of its use during the 1980s. In this sense, ad- ICD: International Classification of Diseases vances in diagnosis and better death certification would have yielded a rise in the period effect. However, trends Funding The study was funded by a research grant from the Spanish Health Research become less pronounced from the 1980s onwards. Fund (FIS PI11/00871). The funding body had no role in the design of the This study has some advantages and limitations that study and collection, analysis, and interpretation of data and in writing the should be taken into account. Among its strengths, it in- manuscript. volves the follow-up of the total Spanish population Availability of data and materials through 60 years. This is a dynamic cohort, with entries Data from the Spanish National Statistics Institute used in this study are and exits across the study period, which encompasses public. For the calendar period 1952 to 1974, national population figures, generations born approximately from 1865 to 1975– together with number of deaths were obtained from the official annual reports of the National Statistics Institute. From 1975 to 2012, data are 1985 and thus constitutes a long time series. Also, available on the Spanish National Statistics Institute website (http:// survival of pancreatic cancer continues to be very low www.ine.es/dyngs/INEbase/listaoperaciones.htm). The datasets used and [2], and therefore, mortality statistics can be consid- analysed during the current study are available from the corresponding author on reasonable request. ered as a good proxy for incidence and are represen- tative of the epidemiology of the disease. On the Authors’ contributions other hand, our results rely on the accuracy of death DSM analyzed and interpreted the data and drafted the manuscript. ON analyzed the data, formatted the figures and revised the manuscript. NFL, certificates and coding practices, and there could have BPG and MP interpreted the results and revised the manuscript. GLA been changes in their quality along the study period. designed the study, analyzed and interpreted the results, formatted the However, cancer death certificates in Spain possess an figures and revised the manuscript. NA designed the study, interpreted the results and drafted the manuscript. All authors read and approved the final accuracy comparable to that reported for other indus- manuscript. trialized countries, and pancreatic cancer is among the cancer sites classified as well certified according Ethics approval and consent to participate to published quality indicators [32]. Not applicable. Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 10 of 10 Competing interests 23. Instituto Nacional de Estadística. Encuesta Nacional de Salud (serie histórica). The authors declare that they have no competing interests. Portal estadístico [Internet]. Ministerio de Sanidad, Servicios Sociales e Igualdad [cited 30/06/2017]. Available from: http://pestadistico. inteligenciadegestion.msssi.es/publicoSNS/Comun/ArbolNodos. Publisher’sNote aspx?idNodo=42. Springer Nature remains neutral with regard to jurisdictional claims in 24. Observatorio Español de la Droga y las Toxicomanías. Estadísticas 2015. published maps and institutional affiliations. Alcohol, tabaco y drogas ilegales en España [Internet]. Ministerio de Sanidad, Servicios Sociales e Igualdad. 2016 [cited 30/06/2017]. Available Author details from: http://www.pnsd.msssi.gob.es/profesionales/sistemasInformacion/ 1 2 Research Unit, Spanish Society of Rheumatology, Madrid, Spain. Cancer and informesEstadisticas/pdf/INFORME_2015.pdf. Environmental Epidemiology Unit, National Center for Epidemiology, Carlos 25. Seoane-Mato D, et al. Trends in oral cavity, pharyngeal, oesophageal and III Institute of Health, Madrid, Spain. Consortium for Biomedical Research in gastric cancer mortality rates in Spain, 1952-2006: an age-period-cohort Epidemiology and Public Health (CIBER Epidemiología y Salud Pública, analysis. BMC Cancer. 2014;14:254. CIBERESP), Madrid, Spain. 26. Shen Q-W, Yao Q-Y. Total fat consumption and pancreatic cancer risk: a meta-analysis of epidemiologic studies. Eur J Cancer Prev. 2015;24:278–85. Received: 14 December 2016 Accepted: 9 May 2018 27. Genkinger JM, et al. Dairy products and pancreatic cancer risk: a pooled analysis of 14 cohort studies. Ann Oncol. 2014;25:1106–15. 28. Behrens G, et al. Physical activity and risk of pancreatic cancer: a systematic References review and meta-analysis. Eur J Epidemiol. 2015;30:279–98. 1. Ferlay, J. et al. GLOBOCAN 2012 v1.0, Cancer incidence and mortality 29. Farris MS, Mosli MH, McFadden AA, Friedenreich CM, Brenner DR. The worldwide: IARC CancerBase No. 11 [Internet]. (International Agency for association between leisure time physical activity and pancreatic Cancer risk Research on Cancer, 2013). in adults: a systematic review and meta-analysis. Cancer Epidemiol Biomark 2. De Angelis R, et al. Cancer survival in Europe 1999-2007 by country and Prev. 2015;24:1462–73. age: results of EUROCARE–5-a population-based study. Lancet Oncol. 2014; 30. Guan H-B, Wu L, Wu Q-J, Zhu J, Gong T. Parity and pancreatic cancer risk: a 15:23–34. dose-response meta-analysis of epidemiologic studies. PLoS One. 2014;9: 3. Instituto de Salud Carlos III. Mortalidad por Cáncer en España (2012). http:// e92738. www.isciii.es/ISCIII/es/contenidos/fd-servicios-cientifico-tecnicos/fd-vigilancias- 31. Zhu B, et al. Parity and pancreatic cancer risk: evidence from a meta-analysis alertas/fd-epidemiologia-ambiental-y-cancer/mortalidad-cancer-en-espana. of twenty epidemiologic studies. Sci Rep. 2014;4:5313. shtml (2014). 32. Pérez-Gómez B, et al. Accuracy of cancer death certificates in Spain: a 4. López-Abente, G., Núñez, O., Pérez-Gómez, B., Aragonés, N. & Pollán, M. La summary of available information. Gac Sanit. 2006;20(Suppl 3):42–51. situación del cáncer en España: Informe 2015. (Centro Nacional de Epidemiología, 2015). 5. Hidalgo M. Pancreatic cancer. N Engl J Med. 2010;362:1605–17. 6. Maisonneuve P, Lowenfels AB. Risk factors for pancreatic cancer: a summary review of meta-analytical studies. Int J Epidemiol. 2015;44:186–98. 7. American Institute for Cancer Research. World Cancer Research Fund. Food, physical activity and the prevention of cancer: a global perspective. (AICR, 2007). 8. Ekbom A, Trichopoulos D. Pancreatic Cancer. In: Hans-Olov Adami, David Hunter, and Dimitrios Trichopoulos. In: Textbook of Cancer Epidemiology. Second ed. Oxford (United Kingdom): Oxford University Press; 2008. 9. Fritschi L, et al. Occupational exposure to N-nitrosamines and pesticides and risk of pancreatic cancer. Occup Environ Med. 2015;72:678–83. 10. Pace, M. et al. Revision of the European standard population. Report of the Eurostat’s task force. (Publications Office of the European Union, 2013). 11. Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med. 2000;19:335–51. 12. Osmond C, Gardner MJ. Age, period and cohort models applied to cancer mortality rates. Stat Med. 1982;1:245–59. 13. Holford TR. Understanding the effects of age, period, and cohort on incidence and mortality rates. Annu Rev Public Health. 1991;12:425–57. 14. Breslow NE. Extra-Poisson variation in log-linear models. Appl Stat. 1984;33: 38–44. 15. Muggeo VM. Estimating regression models with unknown break-points. Stat Med. 2003;22:3055–71. 16. Muggeo VM. Segmented: segmented relationships in regression models. R package version. 2004;0:1–4. 17. Søreide K, Aagnes B, Møller B, Westgaard A, Bray F. Epidemiology of pancreatic cancer in Norway: trends in incidence, basis of diagnosis and survival 1965-2007. Scand J Gastroenterol. 2010;45:82–92. 18. Lefebvre A-C, et al. Pancreatic cancer: incidence, treatment and survival trends–1175 cases in calvados (France) from 1978 to 2002. Gastroentérologie Clin Biol. 2009;33:1045–51. 19. Riall TS, et al. Pancreatic cancer in the general population: improvements in survival over the last decade. J Gastrointest Surg. 2006;10:1212–1223; discussion 1223–1224. 20. Bosetti C, et al. Cancer mortality in Europe, 2005-2009, and an overview of trends since 1980. Ann Oncol. 2013;24:2657–71. 21. Ma J, Siegel R, Jemal A. Pancreatic cancer death rates by race among US men and women, 1970-2009. J Natl Cancer Inst. 2013;105:1694–700. 22. Fung S, Forte T, Rahal R, Niu J, Bryant H. Provincial rates and time trends in pancreatic cancer outcomes. Curr Oncol. 2013;20:279–81. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Cancer Springer Journals

Long-term trends in pancreatic cancer mortality in Spain (1952–2012)

Free
10 pages

Loading next page...
 
/lp/springer_journal/long-term-trends-in-pancreatic-cancer-mortality-in-spain-1952-2012-5dvvQjBdZB
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology; Health Promotion and Disease Prevention; Biomedicine, general; Medicine/Public Health, general
eISSN
1471-2407
D.O.I.
10.1186/s12885-018-4494-3
Publisher site
See Article on Publisher Site

Abstract

Background: Pancreatic cancer is acquiring increasing prominence as a cause of cancer death in the population. The purpose of this study was to analyze long-term pancreatic cancer mortality trends in Spain and evaluate the independent effects of age, death period and birth cohort on these trends. Methods: Population and mortality data for the period 1952–2012 were obtained from the Spanish National Statistics Institute. Pancreatic cancer deaths were identified using the International Classification of Diseases ICD-6 to ICD-9 (157 code) and ICD-10 (C25 code). Age-specific and age-adjusted mortality rates were computed by sex, region and five-year period. Changes in pancreatic cancer mortality trends were evaluated using joinpoint regression analyses by sex and region. Age-period-cohort log-linear models were fitted separately for each sex, and segmented regression models were used to detect changes in period- and cohort-effect curvatures. Results: In men, rates increased by 4.1% per annum from 1975 until the mid-1980s and by 1.1% thereafter. In women, there was an increase of 3.6% per annum until the late 1980s, and 1.4% per annum from 1987 to 2012. With reference to the cohort effects, there was an increase in mortality until the generations born in the 1950s in men and a subsequent decline detected by the change point in 1960. A similar trend was observed in women, but the change point occurred 10 years later than in men. Conclusions: Pancreatic cancer mortality increased over the study period in both sexes and all regions. An important rise in rates -around 4% annually- was registered until the 1980s, and upward trends were more moderate subsequently. The differences among sexes in trends in younger generations may be linked to different past prevalence of exposure to some risk factors, particularly tobacco, which underwentan earlier decrease in menthaninwomen. Keywords: Pancreatic cancer, Tobacco smoking, Mortality, Age-period-cohort analysis, Change-points, Time trends, Spain Background males and 7.6 cases per 100,000 females [1]. According to Though, in terms of incidence, pancreatic cancer is not these data, this cancer has become the tenth most com- among the most frequent cancers, its high lethality mon cancer type registered among men and the sixth places this malignant tumor among those that cause a among women. higher number of deaths worldwide [1]. The overall As regards to mortality, in 2012 pancreatic cancer prognosis of pancreatic cancer is extremely poor, with ranked seventh as cause of cancer death among Spanish five-year relative survival rates around 6% in Europe [2]. men, with age adjusted mortality rates of 10.7 per 100,000 In Spain, 6367 new pancreatic cancer cases were esti- inhabitants [3], being the third leading cause of oncologic mated to occur in 2012, with age adjusted incidence rates deaths among men between 40 and 59 years [4]. In (European standard population) of 11.5 cases per 100,000 women, pancreatic cancer was the fourth most common cause of oncologic death, with rates around 6.8 per 100,000. In sum, pancreatic cancer is currently responsible * Correspondence: nuria.aragones@salud.madrid.org Cancer and Environmental Epidemiology Unit, National Center for for 5 and 7% of the total number of deaths due to cancer Epidemiology, Carlos III Institute of Health, Madrid, Spain among Spanish males and females, respectively. Consortium for Biomedical Research in Epidemiology and Public Health (CIBER Epidemiología y Salud Pública, CIBERESP), Madrid, Spain Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 2 of 10 The etiology of pancreatic cancer is unclear [5]and, as time trends for each region and sex. We used the joinpoint in other cancers, probably multifactorial. Several factors regression analysis to evaluate the presence of change have been suggested as possible causes for this neoplasm, points in adjusted mortality rates over time by sex and re- but their contribution, according to their relative risks, is gion and to estimate the annual percent of change over the small [6]. Between 3 and 7% of cases could be associated study period [11]. Ceuta and Melilla regions were excluded with genetic susceptibility. Regarding exogenous expo- from this analysis, because of their small populations. sures, tobacco smoking (with strong evidence) and Helico- With respect to age-period-cohort models, first, bacter pylori infection (with moderate evidence) have age-specific mortality rates per 100,000 person-years were been considered the major risk factors for pancreatic can- computed by sex and calendar period (using five-year pe- cer, in terms of their population attributable fraction, in a riods) at the national level. Then, separate log-linear Pois- recent review [6]. According to the American Institute for son models were fitted to study the effect of age, period of Cancer Research, there is also convincing evidence to con- death and birth cohort for each sex on mortality trends. sider body fatness as a risk factor for pancreatic cancer To address the “non-identifiability” problem (i.e. the three [7]. Other exposures or clinical entities that have been factors -age, period and cohort- are linearly dependent), suggested to be associated with increased risk include high we used Osmond and Gardner’ssolution[12], as well as red meat consumption, type II Diabetes Mellitus, chronic curvature effects and net drift as proposed by Holford pancreatitis, high alcohol consumption, hepatitis B virus [13]. The Osmond-Gardner solution splits net drift into infection and specific occupational exposures, such as cohort and period slopes, by minimizing any disagreement certain pesticides, organic solvents, polycyclic aromatic in parameter estimates between the full three-factor hydrocarbons and nickel compounds [6–9]. On the other model and each of the two-factor models (age-period, hand, several factors could have a preventive effect, as it is age-cohort and period-cohort) according to their good- the case of allergic conditions, high fruit and vegetable ness of fit. Moreover, it allows to estimate two parameters consumption and physical activity [6–8]. not affected by the non-identifiability problem: (i) overall This study aimed to monitor pancreatic cancer mortal- change over time (denominated net drift), which is the ity trends since the middle of the twentieth century until sum of the cohort and period slopes [13]; and (ii) devi- recent years in Spain, using joinpoint regression models ation of any period or cohort estimates from the general and age-period-cohort analyses to evaluate the inde- trend (denominated curvature). Age groups < 30 years pendent effects of age, death period and birth cohort on were excluded from this analysis due to the limited num- these trends. ber of deaths. The open-ended category of persons aged 85 years and over was also excluded. We checked for Methods extra-Poisson dispersion [14] and, where present, effects Mortality data for the calendar period 1952–2012 were were calculated using a negative binomial distribution. obtained from the Spanish National Statistics Institute The presence of change points and 95% confidence inter- (Instituto Nacional de Estadística) at national level. Dur- vals in the curvatures of the cohort and period effects was ing this period, different revisions of the International evaluated by fitting segmented models to the relationship Classification of Diseases (ICD) have been used. Codes between curvature effects and time. Details of the recursive selected to identify deaths due to pancreatic cancer were algorithm used to estimate the segmented regression have adapted accordingly: code 157 in the ICD-6 to ICD-9 been published elsewhere [15], and the procedure can be and code C25 in the ICD-10. Population data corre- easily fitted using the R package “segmented” [16]. sponding to censuses and municipal rolls for the mid- year of each quinquennium were also obtained from the Results Spanish National Statistics Institute. As a geographic reference, Fig. 1 shows the location of the From 1975 to 2012, mortality and population data are Spanish Autonomous Communities, with the regional dis- public and available at regional level, and were stratified by tribution of pancreatic cancer mortality in both sexes in sex, five-year-age group (from 0 to 4 to 85+ years), calendar the last quinquennium (2008–2012). AAMRs (1976 ESP) year and region (Autonomous Community). Age-adjusted by sex, Autonomous Community and calendar period are mortality rates (AAMR) per 100,000 person-years were presented in Table 1. AAMRs have increased in both sexes then calculated, by the direct method, for each sex, in all regions, but not uniformly. The largest increases oc- five-year calendar period and region, using the 1976 Euro- curred before the 1990s in both sexes. Then, pancreatic pean Standard Population (ESP). Age-adjusted mortality cancer mortality rates have increased at a slower speed. rates were also calculated using the 2013 ESP to allow com- No differences in this trend were observed between men parison with other works (Additional files 1 and 2)[10]. and women. In both sexes, the highest increment took Additionally, annual age-adjusted mortality rates and their place between the 1978–1982 and the 1983–1987 quin- corresponding standard errors were calculated to study quennia (18% in men and 23% in women), with more Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 3 of 10 Asturias Cantabria Basque Country Galicia Navarre La Rioja Catalonia Castile and Leon Aragon Madrid Castile Balearic Islands Valencian La Mancha Community Extremadura Murcia Andalousia [6.11,7.78] (7.78,8.53] (8.53,8.65] Canary Islands Ceuta/Melilla (8.65,8.97] (8.97,9.71] Fig. 1 Pancreatic cancer mortality in Spain (2008–2012): AAMR per 100,000 person-years (1976 ESP) by Autonomous Community modest increments thereafter (between 5 and 11%). Differ- late 1980s, and then increased by 1.4% per annum from ences among regions over the study period have been 1987 to the end of the study period. By region, some of slightly reduced: while in the quinquennium 1978–1982 them also showed a two-phase pattern, although with dif- the ratio between the highest and lowest rates was around ferences in the year when the change was estimated to 1.8 in males and 1.7 in females, in 2008–2012 this ratio was have occurred. Meanwhile, in others no inflection points around 1.3 in both sexes. In men, the highest rates in the were detected and rates increased at the same speed quinquennium 2008–2012 were found in Asturias, La Rioja through the study period. Asturias was the Autonomous and Galicia, whereas in women the highest rates Community with the smaller overall increase (the only corresponded to Navarra, Asturias and Cantabria. Since a one under 1%) in both sexes, though their mortality rates new European Standard Population has been published re- were among the highest of the country in all quinquennia. cently, AAMRs were recalculated using the 2013 ESP Figure 2 shows the evolution of smoothed pancreatic can- (Additional file 1: Figure S1 and Additional file 2:Table S1). cer death rates over time by sex and region. The presence Results are similar, though AAMRs tend to be higher when of changes in trends is visible. using the 2013 ESP, given the greater weight that this new Figure 3 depicts age-specific rates by birth cohort, in standard population gives to older age groups. men and women. In both sexes, clear increases over time The results from joinpoint regression analyses over the are observed in all age groups over 45 years of age. How- period 1975–2012 by Autonomous Community and sex, ever, in males younger than 45 years of age a trend to and in Spain as a whole are presented in Table 2.Bothin stabilization, or even a reduction, is observed, while men and women statistically significant upward trends among women this trend to stabilization is only observed were seen in nearly all Autonomous Communities. In for the 30–35 age group. Nevertheless, trends in the youn- men, there was an overall increase in the rates of 2% per gest age groups are based on a small number of deaths. annum. Joinpoint analysis detected a change point in the Table 3 presents the goodness of fit of the different mid-1980s: during the first period, rates increased by 4.1% age-period-cohort models. In men, period effect was the per annum and by 1.1% during the second period. In main contributor to the estimated trend in mortality, while women, pancreatic cancer mortality rates experienced a in women the main contributor was the cohort effect. marked increase of 3.6% per annum from 1975 until the Figure 4 depicts cohort and period effects with the change Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 4 of 10 Table 1 Pancreatic cancer mortality in Spain (1978–2012) by sex, Autonomous Community and calendar period 1978–82 1983–87 1988–92 1993–97 1998–02 2003–07 2008–12 Men Andalucía 5.34 6.18 6.71 7.37 7.57 8.29 9.34 Aragón 5.91 7.57 9.39 8.95 9.83 10.38 11.17 Asturias 8.81 11.67 10.26 9.79 11.10 11.64 12.37 C.Valenciana 6.33 7.84 7.71 8.45 9.33 9.04 10.60 Cantabria 7.88 9.49 10.78 10.74 10.33 11.79 10.81 Castilla y León 6.43 7.36 8.39 9.38 9.06 10.46 10.59 Castilla-la Mancha 4.86 5.96 6.78 7.18 7.74 8.68 9.26 Cataluña 6.91 8.34 8.95 9.67 9.44 10.07 10.79 Ceuta 5.14 10.31 11.69 14.56 13.21 13.66 9.54 Extremadura 6.18 6.86 9.18 8.86 10.50 9.80 11.36 Galicia 5.90 8.14 8.54 9.70 10.85 10.99 11.47 Islas Baleares 6.57 7.89 9.17 7.99 9.38 9.81 9.57 Islas Canarias 9.26 9.30 10.77 9.66 10.38 10.55 10.21 La Rioja 7.77 8.01 10.43 9.67 9.08 10.46 12.17 Madrid 5.93 5.76 7.99 7.92 9.31 9.52 9.19 Melilla 1.68 2.09 0.87 3.77 8.49 8.34 8.39 Murcia 6.16 6.33 6.48 7.73 9.06 9.66 10.21 Navarra 6.49 8.25 9.89 10.98 11.37 10.87 11.30 País Vasco 7.28 8.89 9.49 10.74 10.42 10.65 10.75 Spain 6.33 7.49 8.31 8.81 9.32 9.77 10.32 Women Andalucía 3.41 4.11 4.37 4.75 4.72 5.02 5.97 Aragón 3.83 4.85 5.14 6.01 5.87 6.33 6.37 Asturias 5.27 6.51 5.80 6.02 6.03 6.71 7.26 C.Valenciana 3.17 4.37 4.75 5.34 5.45 6.03 6.69 Cantabria 4.34 5.44 6.52 8.17 5.97 7.19 7.14 Castilla y León 3.57 4.61 4.74 5.35 5.69 5.77 6.81 Castilla-la Mancha 3.12 3.75 4.87 4.69 4.89 5.75 6.53 Cataluña 3.97 4.80 5.21 5.28 5.88 6.12 6.69 Ceuta 6.39 7.07 7.90 6.99 8.08 6.28 4.56 Extremadura 4.18 4.12 4.92 5.55 5.39 6.15 6.88 Galicia 3.35 4.36 4.50 5.47 5.89 6.16 6.58 Islas Baleares 3.52 4.44 4.95 5.77 5.58 5.98 6.01 Islas Canarias 3.74 5.06 6.24 6.42 6.58 6.91 6.69 La Rioja 4.46 5.57 5.53 5.44 6.95 6.27 6.61 Madrid 3.19 3.48 4.49 5.09 5.61 5.84 6.47 Melilla 3.45 1.25 3.63 5.18 4.28 2.97 2.46 Murcia 4.51 3.59 4.03 5.08 4.99 5.46 6.64 Navarra 3.86 5.28 5.12 6.49 6.69 6.97 8.18 País Vasco 3.86 5.25 5.74 5.81 6.47 6.44 6.80 Spain 3.64 4.46 4.89 5.36 5.61 5.96 6.57 Age-adjusted mortality rates per 100,000 person-years (1976 ESP) Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 5 of 10 Table 2 Pancreatic cancer mortality trend changes in Spain evaluated using joinpoint analysis by sex and Autonomous Community Period 1 Period 2 APC N. of change points Years APC 1 Years APC 2 a a Men Andalucía 2.89 1 1975-1978 17.66 1978-2012 1.68 Aragón 1.71 0- - - - Asturias 0.84 0- - - - a a a C. Valenciana 2.14 1 1975-1985 4.77 1985-2012 1.19 Cantabria 0.84 0- - - - a a a Castilla y León 1.93 1 1975-1990 3.39 1990-2012 0.95 Castilla-la Mancha 2.00 0- - - - a a a Cataluña 1.79 1 1975-1987 3.90 1987-2012 0.80 Extremadura 1.80 0- - - - a a a Galicia 2.82 1 1975-1985 6.93 1985-2012 1.34 a a Islas Baleares 2.21 1 1975-1986 6.51 1986-2012 0.45 Islas Canarias 2.37 1 1975-1978 29.02 1978-2012 0.30 La Rioja 1.38 0- - - - a a Madrid 1.62 1 1975-2001 2.50 2001-2012 -0.42 Murcia 1.74 0- - - - Navarra 1.39 0- - - - País Vasco 1.03 0- - - - a a a Spain 1.99 1 1975-1986 4.10 1986-2012 1.11 Women Andalucía 1.57 0- - - - Aragón 1.71 0- - - - Asturias 0.94 0- - - - a a a C. Valenciana 2.78 1 1975-1984 6.48 1984-2012 1.61 a a Cantabria 2.23 1 1975-1991 5.55 1991-2012 -0.23 Castilla y León 1.89 0- - - - Castilla-la Mancha 2.34 0- - - - Cataluña 1.52 0- - - - Extremadura 1.65 0- - - - a a Galicia 3.70 1 1975-1977 38.72 1977-2012 1.99 Islas Baleares 1.54 0- - - - a a Islas Canarias 1.76 1 1975-1992 3.64 1992-2012 0.18 La Rioja 0.84 0 - - - - a a a Madrid 2.55 1 1975-1994 3.61 1994-2012 1.43 Murcia 1.65 0- - - - Navarra 2.08 0- - - - a a a País Vasco 1.66 1 1975-1989 3.03 1989-2012 0.84 a a a Spain 2.09 1 1975-1987 3.56 1987-2012 1.39 APC Annual Percentage of Change Statistically significant trend as obtained from the segmented regression. Statistical tests were two sided. The significance level was considered as 0.05 points (listed in Table 4) detected on their curvatures. With 1960. In women, as can be seen from its flatter curvature respect to the cohort effect in men, there was an increase (thin line), the cohort effect is less pronounced and the in mortality until the generations born in the 1950s and a change point was placed a decade later, though this subsequent decline with a detectable change point around finding is difficult to assess as it is based on very few Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 6 of 10 Men Women La Rioja Aragon Galicia C.Valenciana Extremadura Castilla y León Murcia Cataluña País Vasco Andalucia Navarra Castilla−la Mancha Asturias Madrid Islas Baleares Cantabria Islas Canarias Navarra Castilla y León Galicia Cantabria Castilla−la Mancha C.Valenciana La Rioja Extremadura Madrid Andalucia Murcia Cataluña Aragon País Vasco Islas Canarias Asturias Islas Baleares 1975 1985 1995 2005 2015 1975 1985 1995 2005 2015 Calendar Period Fig. 2 Pancreatic cancer mortality trends in Spain (1975–2012), by sex and Autonomous Community. Age-adjusted smoothed mortality rates per 100,000 person-years (1976 ESP) Pancreas men Pancreas women 85+ 80− 85+ 75− 80− 70− 75− 65− 70− 60− 65− 55− 60− 50− 55− 50− 45− 45− 40− 40− 35− 35− 30− 30− 25− 1860 1880 1900 1920 1940 1960 1980 1860 1880 1900 1920 1940 1960 1980 Year of birth Year of birth Fig. 3 AAMR per 100,000 person-years for pancreatic cancer by birth cohort and sex, Spain (1952–2012) Age−specific rates x 100,000 Age standardized death rates (per 100,000 person−years) 1e−02 1e+00 1e+02 Age−specific rates x 100,000 1e−02 1e+00 1e+02 Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 7 of 10 Table 3 Goodness of fit in age, period and cohort models for pancreatic cancer mortality by sex, Spain (1952–2012) Degrees of freedom Deviance % of change in deviance Men Age 121 9980 – Age + drift 120 1382 Reference Age + period 110 354 74.4% Age + cohort 100 509 63.2% Age + period + cohort 90 109 92.1% Women Age 121 6094 – Age + drift 120 694 Reference Age + period 110 327 52.9% Age + cohort 100 153 78.0% Age + period + cohort 90 82 88.2% Pancreas men period cohort 1880 1900 1920 1940 1960 1980 2000 2020 Y Year ear Pancreas women period cohort 1880 1900 1920 1940 1960 1980 2000 2020 Y Year ear Fig. 4 Cohort and period effects on pancreatic cancer mortality by sex, Spain (1952–2012). Cohort and period effects (thick lines), curvature (thin lines) and change points in the curvatures (vertical grey lines) 0.2 0.5 1.0 2.0 4.0 0.2 0.5 1.0 2.0 4.0 Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 8 of 10 Table 4 Cohort and period effect curvature change-points on with the different evolution in the period effect between pancreatic cancer mortality by sex, Spain (1952–2012) men and women in the last years, with a lower annual in- Changes in cohort effect crease of rates in men than in women since the 1990s. In Europe, there are geographical differences in pancre- Birth year Birth year (95% CI) (95% CI) atic cancer mortality trends. Mortality rates have increased in the last three decades in countries like France, Germany, Men 1904.6 (1898.7–1910.5) 1960.6 (1958.9–1962.3) Greece, Italy, Romania or Bulgaria, whereas in Sweden, the Women 1894.4 (1890.9–1897.8) 1971.2 (1968.9–1973.6) United Kingdom and Norway they have decreased (in the last two countries only in men). In Denmark, Ireland, Changes in period effect Finland and Holland, mortality in men diminished until Year of death Year of death 1990s and then began to increase [20]. In the United States, (95% CI) (95% CI) white and black people show opposite trends: mortality Men 1963.5 (1961.7–1965.2) 1987.83 (1985.52–1990.14) rates in white men decreased from 1970 to 1995, and have Women 1964.0 (1957.1–1970.9) increased since then; in white women, there was a slight in- Year of birth with significant trend change as obtained from the segmented crease between 1970 and 1984, a stabilization until the late regression analysis of cohort curvatures from the three-factor model Year of death with significant trend change as obtained from the segmented 1990s, and an increase thereafter. On the other hand, in regression analysis of period curvatures from the three-factor model black men and women, rates increased until the late 1980s – early 1990s and have decreased since then [21]. In rates and deaths. For the period effect, there is a Canada, mortality rates between 1992 and 2009 declined in change in the general trend around 1962–1965 in men and remained stable in women [22]. both sexes, and a second change point in men around Understanding the reasons of the increase in pancre- 1988, not visible in women. atic cancer mortality is challenging, given the complex and not well understood etiology of this neoplasm. In Discussion contrast with other cancer locations, like lung or cervical Our results show that pancreatic cancer mortality rates cancer, which are mainly associated with a unique risk increased over the study period in Spain in both sexes factor, pancreatic cancer has been associated with mul- and all regions. In general, this rise was similar in both tiple factors with modest effect sizes and, some of them, sexes and in areas with higher and lower rates compared with high prevalence of exposure in the general popula- to Spain as a whole. In men, age-adjusted mortality rates tion [6]. Probably, differences in latency periods among annually grew on average by 4.1% in the period 1975– these factors may also have a different effect on pancre- 86, and by 1.1% between 1986 and 2012. In women there atic cancer trends, obscuring their specific contribution. was a similar trend, with a bigger increase until the late Among the risk factors for this cancer, the only univer- 1980s (3.6% annually between 1975 and 1987) than in sally accepted one is tobacco consumption [6]. Trends of the years after (1.4% in the period 1987–2012). pancreatic cancer mortality do not resemble those of other The age-period-cohort analysis shows that, taking the tumors strongly related to smoking, such as lung cancer, mean rate for all cohorts as reference, the risk of dying that is decreasing in males and increasing in females in from pancreatic cancer increased in generations born be- Spain and in most developed countries. Nevertheless, the tween 1870 and 1960 in men and women. This similarity slower increase in pancreatic cancer mortality rates in the in the trends could be related to changes in the exposure last decades could be influenced by the decreasing preva- to risk factors linked to birth cohort and shared by both lence of tobacco consumption. In Spain, prevalence of sexes. The rates observed in young men may indicate a smoking shows a decreasing trend in men since the late levelling off in mortality in the most recent generations. eighties (not previous data available), while in women, Among women, this phenomenon is less clear, though the prevalence rose until the late nineties and then slightly de- stabilization of the rates in the youngest age group could creased [23]. Accordingly, the continuous decrease of the indicate that trend might soon parallel that of men. rates in men cohorts born since the 1960s -not so evident For the period effect, there is a change in the general in women-, could be related to the different evolution in trend around 1962–1965 in males and females and a sec- the prevalence of smoking between sexes. Considering ond change point in men around 1988. Though survival 2013 data, smoking prevalence in men over 34 years old in pancreatic cancer is still very low, a slight improvement had fell 9.8 percentage points in the last decade, while in has been described in some countries in the last years women over 34 years old it had not decreased [24]. This [17–19], which could have contributed to the slower in- would be consistent with a role of tobacco exposure in crease in the mortality rates since the 1990s observed in pancreatic cancer risk, but with the existence of other im- our study. The improvement in survival has been reported portant contributing risk factors that would counterbal- to be slightly higher in men [17]. This would be consistent ance the effect of the reduction in tobacco exposure. Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 9 of 10 Another suspected risk factor for pancreatic cancer is Conclusions Helicobacter pylori infection [6]. Again, trends of pan- This study summarizes the trends in pancreatic cancer creatic cancer mortality do not resemble those of gastric mortality in Spain, allowing for a detailed analysis of the cancer, strongly associated with this infection and whose influence of age, period and cohort effects in each sex. mortality rates are decreasing in both sexes. Neverthe- Like in other developed countries, pancreatic cancer mor- less, according to Fig. 4, the higher increase in the risk tality has been increasing over the last decades. However, in cohorts born until the late nineteenth century-early differences have been identified between sexes. In men, twentieth century and the decrease in those born since mortality rates show a stabilization or even a decrease the 1960s or 1970s in men and women, respectively, is since the early 1990s, mainly among post-1960 birth co- congruent with the epidemiology of H. pylori infection. horts, while in women, a stabilization in the trend is ob- Though this effect is not as clear as it is in gastric cancer servable only among the youngest generations (born after [25], which could be explained by a weaker association the late 1970s). These differences may partially mirror the between H. pylori infection and pancreatic cancer, there evolution of some established risk factors for pancreatic seems to be some coincidence in time between both tu- cancer, such as tobacco exposure, in men and women. mors, especially in women. Further research on the causes of pancreatic cancer is Though with limited evidence, other factors have been needed. In the meanwhile, recommendations to reduce suggested to play a role in the etiology of pancreatic can- exposure to preventable risk factors that have been associ- cer. Among proposed risk factors, extensive research has ated with pancreatic cancer, such as tobacco smoking, focused on the role of diet and anthropometric factors obesity, diet and lack of physical activity may serve to re- [26, 27]. As possible protective factors, a medical history duce the impact of this and other chronic and malignant of allergy, the consumption of fruits and vegetables, diseases. physical activity and parity have been suggested [6, 28– 31]. The controversial evidence from epidemiologic Additional files studies, the diverging trends of this broad spectrum of factors and the different prevalence of exposure among Additional file 1 Figure S1. Pancreatic cancer mortality in Spain (2008–2012): AAMR per 100,000 person-years (2013 ESP) by Autonomous regions make it not easy to disentangle the specific role Community. (PDF 529 kb) of each factor in pancreatic cancer trends. Additional file 2 Table S1. Pancreatic cancer mortality in Spain. AAMR A different aspect that could have played a role in the per 100,000 person-years (2013 ESP) by sex, Autonomous Community observed upward trend in pancreatic cancer mortality is and calendar period. (DOCX 30 kb) the introduction of Computerized Tomography in Spain during the late 1970s of the twentieth century, and the Abbreviations AAMR: Age-adjusted mortality rates; ESP: European Standard Population; spreading of its use during the 1980s. In this sense, ad- ICD: International Classification of Diseases vances in diagnosis and better death certification would have yielded a rise in the period effect. However, trends Funding The study was funded by a research grant from the Spanish Health Research become less pronounced from the 1980s onwards. Fund (FIS PI11/00871). The funding body had no role in the design of the This study has some advantages and limitations that study and collection, analysis, and interpretation of data and in writing the should be taken into account. Among its strengths, it in- manuscript. volves the follow-up of the total Spanish population Availability of data and materials through 60 years. This is a dynamic cohort, with entries Data from the Spanish National Statistics Institute used in this study are and exits across the study period, which encompasses public. For the calendar period 1952 to 1974, national population figures, generations born approximately from 1865 to 1975– together with number of deaths were obtained from the official annual reports of the National Statistics Institute. From 1975 to 2012, data are 1985 and thus constitutes a long time series. Also, available on the Spanish National Statistics Institute website (http:// survival of pancreatic cancer continues to be very low www.ine.es/dyngs/INEbase/listaoperaciones.htm). The datasets used and [2], and therefore, mortality statistics can be consid- analysed during the current study are available from the corresponding author on reasonable request. ered as a good proxy for incidence and are represen- tative of the epidemiology of the disease. On the Authors’ contributions other hand, our results rely on the accuracy of death DSM analyzed and interpreted the data and drafted the manuscript. ON analyzed the data, formatted the figures and revised the manuscript. NFL, certificates and coding practices, and there could have BPG and MP interpreted the results and revised the manuscript. GLA been changes in their quality along the study period. designed the study, analyzed and interpreted the results, formatted the However, cancer death certificates in Spain possess an figures and revised the manuscript. NA designed the study, interpreted the results and drafted the manuscript. All authors read and approved the final accuracy comparable to that reported for other indus- manuscript. trialized countries, and pancreatic cancer is among the cancer sites classified as well certified according Ethics approval and consent to participate to published quality indicators [32]. Not applicable. Seoane-Mato et al. BMC Cancer (2018) 18:625 Page 10 of 10 Competing interests 23. Instituto Nacional de Estadística. Encuesta Nacional de Salud (serie histórica). The authors declare that they have no competing interests. Portal estadístico [Internet]. Ministerio de Sanidad, Servicios Sociales e Igualdad [cited 30/06/2017]. Available from: http://pestadistico. inteligenciadegestion.msssi.es/publicoSNS/Comun/ArbolNodos. Publisher’sNote aspx?idNodo=42. Springer Nature remains neutral with regard to jurisdictional claims in 24. Observatorio Español de la Droga y las Toxicomanías. Estadísticas 2015. published maps and institutional affiliations. Alcohol, tabaco y drogas ilegales en España [Internet]. Ministerio de Sanidad, Servicios Sociales e Igualdad. 2016 [cited 30/06/2017]. Available Author details from: http://www.pnsd.msssi.gob.es/profesionales/sistemasInformacion/ 1 2 Research Unit, Spanish Society of Rheumatology, Madrid, Spain. Cancer and informesEstadisticas/pdf/INFORME_2015.pdf. Environmental Epidemiology Unit, National Center for Epidemiology, Carlos 25. Seoane-Mato D, et al. Trends in oral cavity, pharyngeal, oesophageal and III Institute of Health, Madrid, Spain. Consortium for Biomedical Research in gastric cancer mortality rates in Spain, 1952-2006: an age-period-cohort Epidemiology and Public Health (CIBER Epidemiología y Salud Pública, analysis. BMC Cancer. 2014;14:254. CIBERESP), Madrid, Spain. 26. Shen Q-W, Yao Q-Y. Total fat consumption and pancreatic cancer risk: a meta-analysis of epidemiologic studies. Eur J Cancer Prev. 2015;24:278–85. Received: 14 December 2016 Accepted: 9 May 2018 27. Genkinger JM, et al. Dairy products and pancreatic cancer risk: a pooled analysis of 14 cohort studies. Ann Oncol. 2014;25:1106–15. 28. Behrens G, et al. Physical activity and risk of pancreatic cancer: a systematic References review and meta-analysis. Eur J Epidemiol. 2015;30:279–98. 1. Ferlay, J. et al. GLOBOCAN 2012 v1.0, Cancer incidence and mortality 29. Farris MS, Mosli MH, McFadden AA, Friedenreich CM, Brenner DR. The worldwide: IARC CancerBase No. 11 [Internet]. (International Agency for association between leisure time physical activity and pancreatic Cancer risk Research on Cancer, 2013). in adults: a systematic review and meta-analysis. Cancer Epidemiol Biomark 2. De Angelis R, et al. Cancer survival in Europe 1999-2007 by country and Prev. 2015;24:1462–73. age: results of EUROCARE–5-a population-based study. Lancet Oncol. 2014; 30. Guan H-B, Wu L, Wu Q-J, Zhu J, Gong T. Parity and pancreatic cancer risk: a 15:23–34. dose-response meta-analysis of epidemiologic studies. PLoS One. 2014;9: 3. Instituto de Salud Carlos III. Mortalidad por Cáncer en España (2012). http:// e92738. www.isciii.es/ISCIII/es/contenidos/fd-servicios-cientifico-tecnicos/fd-vigilancias- 31. Zhu B, et al. Parity and pancreatic cancer risk: evidence from a meta-analysis alertas/fd-epidemiologia-ambiental-y-cancer/mortalidad-cancer-en-espana. of twenty epidemiologic studies. Sci Rep. 2014;4:5313. shtml (2014). 32. Pérez-Gómez B, et al. Accuracy of cancer death certificates in Spain: a 4. López-Abente, G., Núñez, O., Pérez-Gómez, B., Aragonés, N. & Pollán, M. La summary of available information. Gac Sanit. 2006;20(Suppl 3):42–51. situación del cáncer en España: Informe 2015. (Centro Nacional de Epidemiología, 2015). 5. Hidalgo M. Pancreatic cancer. N Engl J Med. 2010;362:1605–17. 6. Maisonneuve P, Lowenfels AB. Risk factors for pancreatic cancer: a summary review of meta-analytical studies. Int J Epidemiol. 2015;44:186–98. 7. American Institute for Cancer Research. World Cancer Research Fund. Food, physical activity and the prevention of cancer: a global perspective. (AICR, 2007). 8. Ekbom A, Trichopoulos D. Pancreatic Cancer. In: Hans-Olov Adami, David Hunter, and Dimitrios Trichopoulos. In: Textbook of Cancer Epidemiology. Second ed. Oxford (United Kingdom): Oxford University Press; 2008. 9. Fritschi L, et al. Occupational exposure to N-nitrosamines and pesticides and risk of pancreatic cancer. Occup Environ Med. 2015;72:678–83. 10. Pace, M. et al. Revision of the European standard population. Report of the Eurostat’s task force. (Publications Office of the European Union, 2013). 11. Kim HJ, Fay MP, Feuer EJ, Midthune DN. Permutation tests for joinpoint regression with applications to cancer rates. Stat Med. 2000;19:335–51. 12. Osmond C, Gardner MJ. Age, period and cohort models applied to cancer mortality rates. Stat Med. 1982;1:245–59. 13. Holford TR. Understanding the effects of age, period, and cohort on incidence and mortality rates. Annu Rev Public Health. 1991;12:425–57. 14. Breslow NE. Extra-Poisson variation in log-linear models. Appl Stat. 1984;33: 38–44. 15. Muggeo VM. Estimating regression models with unknown break-points. Stat Med. 2003;22:3055–71. 16. Muggeo VM. Segmented: segmented relationships in regression models. R package version. 2004;0:1–4. 17. Søreide K, Aagnes B, Møller B, Westgaard A, Bray F. Epidemiology of pancreatic cancer in Norway: trends in incidence, basis of diagnosis and survival 1965-2007. Scand J Gastroenterol. 2010;45:82–92. 18. Lefebvre A-C, et al. Pancreatic cancer: incidence, treatment and survival trends–1175 cases in calvados (France) from 1978 to 2002. Gastroentérologie Clin Biol. 2009;33:1045–51. 19. Riall TS, et al. Pancreatic cancer in the general population: improvements in survival over the last decade. J Gastrointest Surg. 2006;10:1212–1223; discussion 1223–1224. 20. Bosetti C, et al. Cancer mortality in Europe, 2005-2009, and an overview of trends since 1980. Ann Oncol. 2013;24:2657–71. 21. Ma J, Siegel R, Jemal A. Pancreatic cancer death rates by race among US men and women, 1970-2009. J Natl Cancer Inst. 2013;105:1694–700. 22. Fung S, Forte T, Rahal R, Niu J, Bryant H. Provincial rates and time trends in pancreatic cancer outcomes. Curr Oncol. 2013;20:279–81.

Journal

BMC CancerSpringer Journals

Published: Jun 4, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off