Incidence and mortality from non-Hodgkin lymphoma in Europe: The end of an epidemic?

Authors

  • Cristina Bosetti,

    Corresponding author
    1. Istituto di Ricerche Farmacologiche “Mario Negri”, Via Giuseppe La Masa, 19, 20156 Milan, Italy
    • Istituto di Ricerche Farmacologiche “Mario Negri”, Via Giuseppe La Masa 19, 20156 Milan, Italy
    Search for more papers by this author
    • Fax: +39-0233200231

  • Fabio Levi,

    1. Unité d'épidémiologie du Cancer et Registres Vaudois et Neuchâtelois des Tumeurs, Institut de Médecine Sociale et Préventive (IUMSP), Centre Hospitalier Universitaire Vaudois et Université de Lausanne, Bugnon 17, 1005 Lausanne, Switzerland
    Search for more papers by this author
  • Jacques Ferlay,

    1. International Agency for Research on Cancer, Lyon, France
    Search for more papers by this author
  • Franca Lucchini,

    1. Unité d'épidémiologie du Cancer et Registres Vaudois et Neuchâtelois des Tumeurs, Institut de Médecine Sociale et Préventive (IUMSP), Centre Hospitalier Universitaire Vaudois et Université de Lausanne, Bugnon 17, 1005 Lausanne, Switzerland
    Search for more papers by this author
  • Eva Negri,

    1. Istituto di Ricerche Farmacologiche “Mario Negri”, Via Giuseppe La Masa, 19, 20156 Milan, Italy
    Search for more papers by this author
  • Carlo La Vecchia

    1. Istituto di Ricerche Farmacologiche “Mario Negri”, Via Giuseppe La Masa, 19, 20156 Milan, Italy
    2. Istituto di Statistica Medica e Biometria “G.A. Maccacaro”, Universit̀a degli Studi di Milano, Via Venezian 1, 20133 Milan, Italy
    Search for more papers by this author

Abstract

Non-Hodgkin lymphomas (NHL) are among the few neoplasms whose incidence and mortality have been rising in Europe and North America over the last few decades. To update trends from NHL, we considered mortality data up to 2004 in several European countries, and for comparative purpose in the USA and Japan. We also analyzed patterns in incidence for selected European countries providing national data. In most European countries, NHL mortality rose up to the mid 1990s, and started to level off or decline in the following decade. The rates were, however, still increasing in eastern Europe. Overall, in the European Union, mortality from NHL declined from 4.3/100,000 to 4.1 in men and from 2.7 to 2.5 in women between the late 1990s and the early 2000s. Similarly, NHL mortality rates declined from 6.5/100,000 to 5.5 in US men and from 4.2 to 3.5 in US women. In most countries considered, NHL incidence rates rose up to 1995–99, while they tended to level off or decline thereafter, with particular favorable patterns in countries from northern Europe. Thus, the epidemic of NHL observed during the second half of the 20th century has now started to level off in Europe as in other developed areas of the world. © 2008 Wiley-Liss, Inc.

Non-Hodgkin lymphomas (NHL) are among the few neoplasms whose incidence and mortality rates had been rising for both sexes in Europe and North America since the 1970s.1–4 Over the last 2 decades, NHL mortality rates in the European Union (EU) rose to 4.4/100,000 men and to 2.8/100,000 women (age-standardized, world population). Upward trends were observed also in the USA, whose rates approached 6/100,000 men and 4/100,000 women in the late 1990s, and, although to a smaller extent, in Japan (3.7/100,000 men and 1.9/100,000 women).2

To update trends from NHL, we have considered mortality data up to 2004 in several European countries, and for comparative purpose in the USA and Japan. We have also contrasted trends in incidence with those in mortality for 11 European countries providing national incidence data.

Material and methods

Official death certification data from NHL for 29 European countries, plus the USA and Japan, for the period 1980–2004 were derived from the WHO database available on electronic support.5 Data for Croatia, the Czech Republic and Slovenia were available only since 1985; for Estonia, Lithuania and Slovakia since 1990; for Latvia, Poland, Romania and the Russian Federation since 1995. For Belgium data were available only up to 1997; for Denmark up to 2001; for Italy up to 2002; and for Portugal up to 2003.

The European Union (EU 27) was defined as the 27 member states as since January 2007 (i.e., Austria, Bulgaria, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Lithuania, Luxembourg, Malta, The Netherlands, Portugal, Slovakia, Slovenia, Spain, Sweden, UK), with the exclusion of Cyprus (no data available), Belgium, Latvia Poland and Romania (which provided data only for a limited number of years in the period considered).

In the 2 decades considered, 3 different Revisions of the International classification of diseases (ICD) were used.6–8 NHL were coded as 200, 202, 208, 209 in the 8th ICD Revision, as 200, 202 in the 9th ICD Revision and as C82-85 and C96 in the 10th ICD Revision. There were, however, no major differences in classification and grouping of NHL across the 3 ICD Revisions.

Estimates of the resident population, based on official censuses, were obtained from the same WHO database.5 From the matrices of certified deaths and resident population, age-specific rates for each 5-year age group and calendar period were computed. Age-standardized rates per 100,000, at all ages and truncated at age 35–64 years, were computed using the direct method, based on the world standard population.9 In a few countries, data were missing for part of 1 or more calendar years. No extrapolation was made for missing data.

Incidence data for 11 European countries providing national data at least since the early 1990s onwards were extracted from the successive volumes of Cancer Incidence in Five Continents.10 Additional more recent information was found in various national cancer registry reports available in Internet.11–17

Results

Table I shows the age-standardized mortality rates from NHL per 100,000 men and women in selected European countries, the USA and Japan, in 1990–94, 1995–99 and 2000–04, and the corresponding change in rates. NHL mortality rose up to the mid 1990s in most European countries, while it started to decline in many European countries over the last decade considered. The rates were, however, still increasing in countries from eastern Europe, such as Croatia, Estonia, Lithuania, Romania, the Russian Federation, Slovakia and Slovenia. Changes for most smaller countries were not statistically significant and should be interpreted cautiously, since they can be at least partly explained by random variation. Overall, in the EU, mortality from NHL declined from 4.3/100,000 to 4.1 in men and from 2.7 to 2.5 in women between the late 1990s and the early 2000s. Similarly, NHL mortality rates declined from 6.5/100,000 to 5.5 in US men and from 4.2 to 3.5 in US women, respectively. Corresponding values were 3.9 and 3.7 in Japanese men, and 1.98 and 1.99 in Japanese women.

Table I. Overall Age-Standardized (World Population) Mortality Rates from Non-Hodgkin Lymphoma per 100,000 Men and Women in Selected European Countries, Plus USA and Japan1 in 1990–94, 1995–99 and 2000–04 (Unless Mentioned in Parentheses), and Corresponding Change in Rates
 MenWomen
1990–941995–992000–04% Change1990–941995–992000–04% Change
1997/19922002/19971997/19922002/1997
  • 1

    p < 0.05. Test of significance of the annual percent change in rates using log-linear model based on single calendar years.

  • 2

    Belgium, Latvia, Poland and Romania are not included.

Austria3.433.454.050.617.412.202.362.557.38.1
Bulgaria2.071.721.73−16.910.61.110.890.91−19.82.2
Croatia2.843.413.7220.119.11.711.982.2315.812.6
Czech Republic4.003.503.44−12.51−1.72.192.142.24−2.34.7
Denmark (2000–1)5.644.934.48−12.6−9.13.563.163.10−11.2−1.9
Estonia (1994)2.283.404.4949.132.11.681.932.1614.911.9
Finland5.135.755.2012.11−9.613.643.943.468.2−12.21
France4.384.684.316.81−7.912.702.812.594.1−7.8
Germany3.734.023.687.81−8.512.362.582.409.31−7.0
Greece1.821.671.49−8.2−10.810.961.120.9916.7−11.6
Hungary4.084.033.92−1.2−2.72.302.452.216.5−9.8
Ireland (1990–93)5.145.264.992.3−5.13.913.413.61−12.85.9
Italy (2000–02)4.504.814.676.91−2.92.813.102.9910.31−3.5
Latvia (1996–99)2.663.51 32.01.371.63 19.01
Lithuania (1993–94)2.542.212.74−13.024.01.141.341.5617.5116.4
Luxembourg4.375.183.4918.5−32.63.072.082.64−32.226.9
Malta2.825.564.6897.21−15.83.162.882.19−8.9−24.0
The Netherlands5.175.094.93−1.5−3.13.243.373.144.0−6.8
Norway4.895.144.565.1−11.313.333.462.963.9−14.5
Poland (1995–96, 1999)3.153.181.01.771.801.7
Portugal (2000–03)3.363.813.7513.4−1.62.062.312.3912.113.5
Romania2.272.5311.51.141.3720.21
Russian Federation1.722.1323.810.841.1334.51
Slovakia (1994)1.933.083.1359.611.61.311.671.8727.512.01
Slovenia3.203.293.712.812.82.012.352.3316.9−0.9
Spain3.443.683.567.01−3.32.042.392.2917.21−4.2
Sweden5.145.014.30−2.5−14.213.103.082.82−0.6−8.4
Switzerland6.254.414.36−29.41−1.13.773.032.68−19.61−11.6
UK5.035.123.891.8−24.03.243.312.052.2−38.1
European Union24.174.304.053.11−5.812.592.722.545.01−6.61
USA6.336.525.503.0−15.614.024.163.453.5−17.11
Japan3.633.853.726.11−3.41.911.991.984.21−0.5

The patterns of NHL mortality in middle-aged populations (35–64 years) were similar—and, if anything, more favorable—to those for all ages in most European countries, as in USA and Japan (Table II).

Table II. Overall Age-Standardized (World Population) Mortality Rates from Non-Hodgkin Lymphoma Per 100,000 Men and Women Aged 35–64 Years in Selected European Countries in 1990–94, 1995–99 and 2000–04 (Unless Mentioned in Parentheses), and Corresponding Change in Rates
 MenWomen
1990–941995–992000–04% Change1990–941995–992000–04% Change
1997/19922002/19971997/19922002/1997
  • 1

    p < 0.05. Test of significance of the annual percent change in rates using log-linear model based on single calendar years.

  • 2

    Belgium, Latvia, Poland and Romania are not included.

Austria4.424.374.55−1.14.12.333.032.8430.0−6.3
Bulgaria3.122.862.37−8.3−17.111.671.281.18−23.4−7.8
Croatia3.575.085.0042.31−1.62.282.413.165.731.11
Czech Republic5.534.924.17−11.01−15.22.742.562.59−6.61.2
Denmark (2000–1)6.566.396.09−2.6−4.74.454.534.441.8−2.0
Estonia (1994)2.753.665.9633.162.82.772.343.02−15.529.1
Finland6.297.786.0723.71−22.014.555.023.8410.3−23.51
France5.615.965.086.21−14.813.323.272.83−1.5−13.5
Germany4.845.014.493.5−10.413.013.172.775.3−12.6
Greece2.622.341.80−10.71−23.111.211.381.1914.0−13.8
Hungary5.705.635.41−1.2−3.92.813.282.9116.7−11.3
Ireland 1990–93)6.817.876.7015.61−14.96.214.984.51−19.8−9.4
Italy (2000–02)6.186.295.841.8−7.23.603.753.404.2−9.3
Latvia (1996–99)3.154.62 46.71.862.38 28.0
Lithuania (1993–94)3.832.943.48−23.218.41.461.911.8930.81−1.0
Luxembourg4.568.893.4995.0−60.72.792.993.497.216.7
Malta4.178.325.1899.51−37.74.403.492.47−20.7−29.2
The Netherlands6.376.446.151.1−4.514.144.123.90−0.5−5.3
Norway6.147.185.9816.91−16.715.095.124.190.6−18.2
Poland (1995–96, 1999)4.804.66−2.912.642.45−7.2
Portugal (2000–03)4.985.295.116.2−3.42.793.143.2112.52.2
Romania4.024.112.21.812.26+24.91
Russian Federation2.773.42 23.511.251.68+34.41
Slovakia (1994)3.154.494.5142.50.41.962.512.6828.16.8
Slovenia4.813.963.97−17.70.32.203.022.5637.31−15.2
Spain5.055.154.732.0−8.22.803.082.7910.01−9.4
Sweden6.676.564.56−1.6−30.513.883.632.99−6.4−17.6
Switzerland5.975.224.86−12.61−6.913.943.523.04−10.7−13.6
UK7.207.185.01−0.3−30.24.544.582.490.9−45.6
European Union25.585.705.042.2−11.613.333.413.022.4−11.41
USA8.608.526.42−0.9−24.614.924.913.80−0.2−22.61
Japan4.484.614.162.9−9.812.352.432.393.4−1.6

Figure 1 gives the trends in overall mortality from NHL in the EU, in the USA and Japan. Mortality rates steadily increased in the USA up to the late 1990s, and started to decline thereafter. A similar trend has been observed in the EU, although the decline in more recent calendar years has been less marked. Rates were much lower in Japan, and some leveling of rates was apparent over the most recent calendar years particularly in men.

Figure 1.

Trends in age-standardized (world population) mortality rates from non-Hodgkin lymphoma per 100,000 men and women in the European Union (EU), USA and Japan.

Trends in NHL mortality were similar in countries from northern, western and southern Europe (Fig. 2). The pattern in NHL mortality was less clear for eastern Europe. Since, however, the 3 larger eastern European countries (i.e., Poland, Romania and the Russian Federation) were not included in this area, this pattern should cautiously be interpreted.

Figure 2.

Trends in age-standardized (world population) mortality rates from non-Hodgkin lymphoma per 100,000 men and women in various European areas. North: Denmark, Finland, Ireland, Norway, Sweden, UK; West: Austria, France, Germany, Luxembourg, The Netherlands, Switzerland; South: Croatia, Greece, Italy, Malta, Portugal, Spain; East: Bulgaria, Czech Republic, Estonia, Hungary, Lithuania, Slovakia, Slovenia.

Table III shows the age-standardized incidence rates (overall and at age 35–64 years) from NHL in the 11 European countries providing also incidence data in 1990–94, 1995–99 and 2000–04. In almost all countries considered, NHL incidence rates rose up to the late 1990s, while they tended to level off or decline thereafter. Incidence trends were more favorable in countries from northern Europe, such as Denmark, Finland, The Netherlands and Sweden, particularly in more recent years.

Table III. Overall and Truncated 35 to 64 Age-Standardized (World Population) Incidence Rates from Non-Hodgkin lymphomas Per 100,000 Men and Women in Selected European Countries in 1990–94, 1995–99 and 2000–04 (Unless Mentioned in Parentheses), and the Corresponding Change in Rates
 MenWomen
1990–941995–992000–04% Change1990–941995–992000–04% Change
1997/19922002/19971997/19922002/1997
  • 1

    p < 0.05. Test of significance of the annual percent change in rates using log-linear models based on single calendar years.

All ages
 Czech Republic (2000–03)6.947.737.9311.412.63.694.585.4716.812.2
 Denmark (2000–03)10.0010.359.953.5−3.94.816.967.417.9−1.3
 Finland11.2212.1112.077.91−0.35.578.358.878.31−1.9
 Ireland (1994)9.1210.3511.0813.517.17.758.81−3.217.51
 The Netherlands (2000–03)10.7710.7011.10−0.63.77.047.522.74.01
 Norway10.0410.7110.936.72.14.957.178.119.213.6
 Slovakia (2000–03)5.386.545.9721.61−8.72.583.674.5826.21−1.1
 Slovenia6.817.797.7714.4−0.33.704.755.9220.413.5
 Sweden11.2110.9110.74−2.7−1.65.437.127.412.91.1
 UK, England9.9010.1111.032.119.114.326.637.875.9112.11
 UK, Scotland10.1011.2511.2511.40.05.727.768.369.51−1.6
35–64 years
 Czech Republic (2000–03)10.2011.4911.2512.61−2.17.018.068.2815.012.7
 Denmark (2000–03)15.6016.3816.515.00.811.6012.9812.2211.9−5.9
 Finland17.3018.4717.656.81−4.413.7314.0213.642.1−2.7
 Ireland (1994)13.5617.6318.4630.04.713.1012.4614.28−4.914.6
 The Netherlands (2000–03)16.4916.4516.60−0.20.911.1711.1612.39−0.111.01
 Norway16.3617.5617.927.32.112.4113.6813.4310.2−1.8
 Slovakia (2000–03)8.3010.259.0523.51−11.75.897.047.6719.518.91
 Slovenia10.5710.4510.36−1.1−0.96.648.008.5320.516.6
 Sweden17.0416.2215.43−4.81−4.9110.7810.911.241.13.1
 UK, England15.8516.0316.951.15.7110.9411.5712.715.819.91
 UK, Scotland16.4817.0617.633.53.313.0013.8712.956.7−6.6

Figures 3 and 4 compare trends in age-standardized mortality and incidence rates from NHL in men and women from 11 European countries providing also incidence data. A consistent increase in mortality as well as in incidence was observed in most countries over the last decades, with a tendency, however, to level off or even to decline in more recent years, particularly in countries from northern Europe.

Figure 3.

Trends in age-standardized (world population) mortality and incidence rates from non-Hodgkin lymphoma per 100,000 men in 11 northern and eastern European countries between 1980 and 2004. (+--+ incidence; +—+ mortality).

Figure 4.

Trends in age-standardized (world population) mortality and incidence rates from non-Hodgkin lymphoma per 100,000 women in 11 northern and eastern European countries between 1980 and 2004. (□---□ incidence; □—□ mortality).

Discussion

The present updated analysis of trends from NHL shows that, after the steady increases up to the mid/late 1990s observed in most countries of the EU, mortality started to level off or even decline over more recent years. A similar—although more marked—pattern was observed in the USA.1, 3, 18

NHL incidence has been also consistently upwards until the mid/late 1990s in all European countries providing data, as reported by other time-trend analyses.19, 20 A slowing down in incidence rates was, however, observed over the last years, particularly in Nordic countries,19, 21 as well as in the USA.1, 3, 18

Diagnosis of NHL has improved over time, and these improvements may at least partly be responsible for the observed trends in NHL incidence.22 The lack of upward trends in incidence in most Nordic countries may indeed reflect the fact that in those countries diagnosis and certification of lymphoid neoplasms has long been accurate.23 This line of reasoning is also suggested by the higher lymphoid incidence rates in Nordic countries in earlier calendar periods. Moreover, the complex and evolving classification of lymphomas and the changes in their registration24–26 may have also had some role in NHL (incidence) trends. It is, however, unlikely that advancements in diagnosis and certification accuracy can largely, or totally, account for the earlier generalized rises observed in European countries, as well as other areas of the world.27

Thus, the steady rises registered in NHL rates up to the late 1990s almost certainly reflect real increases in disease incidence. The reasons of the long and steady rise of NHL during the second half of the 20th century, as well as for the more recent leveling off in NHL rates, remain, however, poorly understood. While a few risk factors for selected types of NHL have been identified [including conditions related to the immune system and immunosuppression, specific infectious agents, such as human immunodeficiency virus, HIV and hepatitis C virus (HCV)], the causes of most NHL are largely unknown.28 The subsequent phases of the AIDS/HIV epidemic, including the improvements in survival over recent calendar years, are likely to explain only a limited proportion of NHL trends.29, 30 Similarly, the spread of HCV should have had a limited contribution in the observed rises in NHL, since the fraction of NHL cases attributable to HCV has been relatively low (about 10%) even in countries with a high prevalence of HCV infection, and has been negligible in northern European countries, though not in middle-age men from the USA.31–33 A modest contribution should also have been given by the changed frequency of blood transfusions and transplantations.34

Among other potential risk factors for NHL there are exposure to ultraviolet light solvents and pesticides, but their role in NHL etiology is still undefined and is in any case far from quantified.28 The role of diet on NHL is also open to discussion, but vitamin D and folate, and their recent increase of intake through supplementation, may have some beneficial impact, and be related to the recent favorable trends.35, 36

NHL are much less amenable to treatment than Hodgkin lymphoma, but improvements in survival have been observed in the last few years,37 as a consequence of advancements in treatment for some types of lymphomas, and may have had some role in the recent declines in NHL mortality. Treatments for NHL include radiotherapy for early-stage diseases (indolent NHL) and single-agent or multiagent chemotherapy for advanced-stage diseases.3, 26, 38 More recently, treatments using monoclonal antibodies and autologous bone marrow transplantation were found to improve the survival of recurrent, aggressive NHLs.39, 40

Although histological types of lymphomas affect NHL prognosis,26, 41, 42 they are not available on death certification. The similar pattern of trends in NHL mortality across age groups (characterized by different NHL histotypes), however, weigh against a major role of changes in histological subtypes on recent mortality trends.

Thus, the key message of this comprehensive overview of trends in NHL is that the epidemic of NHL observed during the second half of the 20th century has now started to level off in Europe as in other developed areas of the world.

Acknowledgements

The work of this article was undertaken while C.L.V. was a senior fellow at the International Agency for Research on Cancer. The authors thank Mrs. I. Garimoldi for editorial assistance.

Ancillary