Elsevier

The Lancet Oncology

Volume 18, Issue 8, August 2017, Pages 1022-1039
The Lancet Oncology

Articles
Burden and centralised treatment in Europe of rare tumours: results of RARECAREnet—a population-based study

https://doi.org/10.1016/S1470-2045(17)30445-XGet rights and content

Summary

Background

Rare cancers pose challenges for diagnosis, treatments, and clinical decision making. Information about rare cancers is scant. The RARECARE project defined rare cancers as those with an annual incidence of less than six per 100 000 people in European Union (EU). We updated the estimates of the burden of rare cancers in Europe, their time trends in incidence and survival, and provide information about centralisation of treatments in seven European countries.

Methods

We analysed data from 94 cancer registries for more than 2 million rare cancer diagnoses, to estimate European incidence and survival in 2000–07 and the corresponding time trends during 1995–2007. Incidence was calculated as the number of new cases divided by the corresponding total person-years in the population. 5-year relative survival was calculated by the Ederer-2 method. Seven registries (Belgium, Bulgaria, Finland, Ireland, the Netherlands, Slovenia, and the Navarra region in Spain) provided additional data for hospitals treating about 220 000 cases diagnosed in 2000–07. We also calculated hospital volume admission as the number of treatments provided by each hospital rare cancer group sharing the same referral pattern.

Findings

Rare cancers accounted for 24% of all cancers diagnosed in the EU during 2000–07. The overall incidence rose annually by 0.5% (99·8% CI 0·3–0·8). 5-year relative survival for all rare cancers was 48·5% (95% CI 48·4 to 48·6), compared with 63·4% (95% CI 63·3 to 63·4) for all common cancers. 5-year relative survival increased (overall 2·9%, 95% CI 2·7 to 3·2), from 1999–2001 to 2007–09, and for most rare cancers, with the largest increases for haematological tumours and sarcomas. The amount of centralisation of rare cancer treatment varied widely between cancers and between countries. The Netherlands and Slovenia had the highest treatment volumes.

Interpretation

Our study benefits from the largest pool of population-based registries to estimate incidence and survival of about 200 rare cancers. Incidence trends can be explained by changes in known risk factors, improved diagnosis, and registration problems. Survival could be improved by early diagnosis, new treatments, and improved case management. The centralisation of treatment could be improved in the seven European countries we studied.

Funding

The European Commission (Chafea).

Introduction

The RARECARE project1 defined rare cancers as those with an annual incidence of less than six per 100 000 people in the European Union (EU), and showed that about one in five cancers diagnosed in the EU were rare and slightly more than 4 million people with a rare cancer diagnosis were living in the EU population. Because of their low numbers, the almost 200 rare cancers listed by RARECARE pose challenges for diagnosis, treatments, and clinical decision making. Clinical trials into such cancers are rare too, and it is hard to build up new knowledge and expertise.

The broad consensus suggests that diagnostic pathological confirmation and primary treatment of rare cancers, in particular, should be centralised in reference centres, collaborative networks, or both, with multidisciplinary approaches2 and very specific expertise. Additionally, clinical and translational research calls for a high level of centralisation and international collaboration. To what extent appropriate policies for patients with rare cancer are implemented at the country level has seldom been studied. As a consequence, information for policy makers and stakeholders is scarce for many of these rare tumours.

The project Information Network on Rare Cancers (RARECAREnet) is designed to update epidemiological information about rare cancers in the EU,3 to provide indicators at the country level, time trends, and to study to what extent treatment is centralised in Europe.

In this study, we provide current incidence and survival estimates based on data collected from 94 population-based cancer registries, for 198 rare cancers diagnosed in 2000–07 and for 12 major families of rare cancers. We also present data for the levels of centralisation for rare cancers in selected European countries.

Section snippets

Patients

We extracted data from two databases. The first, the descriptive analysis database, is a subset of the EUROCARE-5 database,4 which includes incidence and follow-up data provided by European population-based cancer registries for patients with cancer diagnosed between Jan 1, 1978, and Dec 31, 2007. Vital status was updated to Dec 31, 2008. From the 117 cancer registries participating in EUROCARE-5, we excluded specialised paediatric cancer registries, the Swedish and Turin cancer registries,

Results

Survival estimates in 2000–07 for all rare cancers were based on 1 994 346 diagnoses, observed by 94 cancer registries. 1 984 147 rare cancer diagnoses were considered for incidence estimates in 2000–07, collected by 83 cancer registries from 1566 million person-years of observation. 62 828 (3%) of 1 984 147 cases were identified by death certificate only or incidentally at autopsy ranging from a maximum of 4412 (10%) of 44 755 cases in Slovakia, to 265 (<1%) of 60 377 cases in Scotland. Data

Discussion

Rare cancers make up one quarter of all malignancies. They are a very heterogeneous group of almost 200 cancers, mostly solid, constituting from 2% of all skin cancers up to 32% of all female genital cancers. We confirmed that 5-year survival is lower for rare cancers than common cancers (49% vs 63%), and for all rare cancer families compared with their more common counterparts, except thoracic cancers. The disadvantage persisted even after excluding common cancers with good prognosis, such as

References (28)

  • Information Network on Rare Cancers

  • S Rossi et al.

    The EUROCARE-5 study on cancer survival in Europe 1999–2007: database, quality checks and statistical analysis methods

    Eur J Cancer

    (2015)
  • Information Network on Rare Cancers. Indicators

  • F Ederer et al.

    The relative survival: a statistical methodology

    Natl Cancer Inst Monogr

    (1961)
  • Cited by (0)

    Members of the RARECARENet Working Group are listed in the appendix

    View full text