1 Istituto Nazionale per lo Studio e la Cura dei Tumori, Epidemiology Unit, Milan; 2 Istituto Superiore di Sanità, Laboratorio di Epidemiologia e Biostatistica, Rome; 3 Istituto Nazionale per lo Studio e la Cura dei Tumori, Cancer Registry Unit, Milan, Italy
Members of the EUROPREVAL Working Group are listed in the Acknowledgements.
*Correspondence to: Dr G. Gatta, Istituto Nazionale per lo Studio e la Cura dei Tumori, Epidemiology Unit, Via Venezian 1, 20133 Milan, Italy. Tel: +39-02-23903518; Fax: +39-02-23903522; Email: gatta{at}istitutotumori.mi.it
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients and methods: Prevalence by year since diagnosis was estimated from incidence and vital status data on 243 471 colon cancer cases collected by EUROPREVAL from 36 European population-based cancer registries. The proportions of cured and fatal cases were estimated by applying cure survival models to the dataset. The proportion of recurrence-free cases was estimated by analysis of a representative sample of 278 colon cancer patients from the Lombardy Cancer Registry (LCR), northern Italy.
Results: The proportions of total prevalence requiring initial care was estimated at 12% in the LCR and 10% in Italy and Europe. Recurrence-free patients formed 89% of the total prevalence in the LCR and 91% in Italy and Europe. Eleven per cent (LCR) and 9% (Italy, Europe) of the total prevalence had recurred and consisted of patients in the terminal phase of their illness.
Conclusions: In 1992, 660 000 people were living with a diagnosis of colon cancer in Europe. We have estimated the proportions of this prevalence requiring particular types health care in the years following diagnosis, providing data useful for planning the allocation of health-care resources.
Key words: care prevalence, colon cancer, Europe, population-based cancer registries
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Time from diagnosis is a major criterion for categorizing people with cancer, since care and surveillance requirements vary with time. In the first few months after diagnosis, care generally consists of primary and adjuvant treatment. Subsequently, patients require follow-up to monitor recurrences or for treatment of side-effects. A subset of cases will require treatment for recurrences, or palliation; thus, presence or absence of recurrence is another useful criterion for subdividing the prevalent population according to the intensity of care required.
For the purposes of planning health resource allocation, it is useful to be able to estimate the prevalences of cancer patients who will continue to need treatment, and those who will require follow-up and possible treatment for their cancer, and to exclude those who will survive for a long time after diagnosis and can be considered cured [1, 2
].
In this paper we propose a method for estimating the prevalences of subgroups of cancer patients, each characterized by fairly homogeneous disease status and health-care requirements. We apply this method to the colon cancer prevalence observed in the population of the province of Varese, northern Italy, covered by the Lombardy Cancer Registry (LCR) [3]. Making certain assumptions, we provide projections of subgroup prevalence for the pool of European colon cancer patients diagnosed in the countries participating in EUROPREVAL [4
]. To do this, we analyzed data on colon cancer from samples of patients archived by the LCR and by 36 European cancer registries participating in EUROPREVAL to obtain relative survival data. We then estimated subgroup prevalences by applying cure models to the relative survival [5
], and by inferring the morbid and pre-morbid prevalence from the total survival and disease-free survival curves [1
], as further explained below.
![]() |
Patients and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
We used three criteria to divide prevalent cases into homogeneous subgroups in terms of need for health-care resources. The first was time from diagnosis: we obtained the distribution of prevalent cases by time from diagnosis from the incidence and follow-up data used to calculate prevalence. The second criterion was cure: we estimated the proportion of patients cured of their cancer by the treatment (cured prevalence). By definition, cured patients have the same life expectancy as individuals in the general population of the same age. Patients who are not cured are those who will eventually die of their cancer; we designated these as the fatal prevalence. The third criterion was recurrence. The recurrence-free prevalence consists of those are cured (cured prevalence) and those who are clinically free of cancer after primary treatment, but later develop recurrence; the latter we designated as the pre-morbid prevalence. In contrast, the proportion of prevalent cases with metastases at diagnosis, or whose disease recurred after shortly primary treatment, we defined as the morbid prevalence.
Clearly, the three criteria are interrelated. For colon cancer in particular, most cases alive 5 or 6 years after diagnosis can be considered cured. Furthermore, almost all recurrent cases are also fatal cases, although a few patients are successfully treated for their recurrence and become long-term survivors.
We designate SO(t) and SR(t) as the observed and relative survival, respectively, of patients t years after diagnosis, and SE(t) as the expected survival of the population as a whole. Some patients are cured of cancer, and therefore have the SE of the general population. In general, we cannot identify individual patients who are cured, but we can estimate the cured prevalence P(0) at the time of diagnosis, i.e. the proportion of the total who will be cured by the treatment. The proportion of cured patients P(t) of the total survivors increases with time, at t years after diagnosis is given by:
![]() |
![]() |
![]() |
Turning now to the recurrent prevalence, we designate SF(t) as the recurrence-free survival function, i.e. the cumulative proportion of patients alive with no clinical manifestation of recurrence t years after diagnosis. This proportion will be lower than, or at best equal to SO(t). By estimating SF(t) from follow-up studies of patients, we obtain the following:
![]() |
![]() |
Figure 1 shows the observed and recurrence-free survival of our patient sample from diagnosis to 11 years later. The figures above the lines are the percentages SF(t)/SO(t) of cases without metastasis/recurrence each year after diagnosis. This proportion increased from 71% at diagnosis to
95% 7 years later. Note that this proportion did not reach 100%, implying that
5% of recurrent cases were alive, and thus probably cured, many years after their recurrence. Ninety-five per cent confidence intervals for the recurrence-free prevalence (lighter lines above and below the solid line) are also shown in Figure 1
; these were calculated using the binomial distribution.
|
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Prevalence breakdowns can be obtained from survey studies on people with cancer; however, surveys are expensive and difficult to conduct, and subject to bias. The breakdowns presented in this paper are population-based, being derived from incidence and follow-up data provided by European cancer registries. However, the disease-free survival estimates were extrapolated from those of a single cancer registry (the LCR) over a limited period (cases diagnosed in 1990) and applied to all Europe. This extrapolation, necessary since data on cancer recurrences are not routinely collected by cancer registries, may be a source of both random variability and bias.
As a measure of random variability we estimated the sensitivity of our prevalence estimates to the random variability of the proportions of recurrence-free patients in the LCR population. We found that most of our prevalence components were fairly insensitive to the random variability of the Varese data, the exception being the recurrent disease prevalence.
Furthermore, to calculate confidence intervals for the proportion of recurrence-free patients we used the standard error of the estimate. This led to rather wide estimates. For example, 5 years after diagnosis the 95% confidence interval for the proportion of recurrence-free patients was 86% to 96%, the lower limit of which is not plausible from a clinical point of view.
With regard to bias, this may occur if different populations have different proportions of fatal cases, different times to recurrence after diagnosis, or different times to death after recurrence. Unfortunately, few data are available to determine whether such differences are important in the populations considered in this study. A previous analysis of colon cancer patients in European countries [5] estimated similar proportions of fatal cases (60% and 61%, respectively) and similar mean survival times of fatal cases (1.13 and 1.29 years) in Italy and Europe in 19831985. Because of the small population size, no similar estimates exist for LCR colon cancer patients. However, there is no evidence from published analyses [11
] of major differences in proportions of cured cases between Varese registries, Italian registries as a whole, and European registries as a whole (5-year relative survival, an approximate measure of cure rate, was 53%, 52% and 50%, respectively).
A previous European study on the incidence of colorectal cancer in 19901991 indicated that 7793% of total incident cases underwent surgery with curative intent, and that 2530% were patients with Dukes' stage C and B2 at diagnosis who should have received adjuvant chemotherapy [12]. In this study we have estimated that 10% of the total European prevalence at that time consisted of cases in initial care (Table 2
).
We also estimated two important fractions of the total prevalence: the morbid prevalence and the pre-morbid prevalence (9% and 5%, respectively, of the total prevalence). The morbid prevalence was estimated from the difference between total survival and disease-free survival. Disease-free survival is usually only available from clinical series; in this study we obtained this information from a population-based incidence sample. For this sample, the observed and disease-free survival curves converged 6 years after diagnosis (37.1% and 33.9%, respectively), indicating that the survivors at this time constitute the fraction of cured patients. This convergence also indicates that the selected sample had adequate follow-up for recurrences, since if recurrences were missed, the curves would have remained more widely separated. The persistence of a small difference between the two curves means that a small fraction of recurrent cases was cured. The sensitivity analysis showed that our estimate of recurrent prevalence was particularly sensitive to random variability in the LCR sample, and must therefore be considered as approximate only.
We estimated that 86% of the total prevalence consisted of those who were cured of their disease. Nevertheless, these patients still require care for the physical and psychological sequelae of treatment. Cured patients have to be followed because it cannot be known in advance whether or not they are cured. For colon cancer, most recurrences occur within 2 years of surgery, and almost all occur within 5 years [13]. This is why colon surveillance programs generally last for 5 years after primary curative treatment, and are intensive in the first 2 years. Nevertheless, surveillance protocols have rarely been subjected to formal efficacy assessment [1
, 14
].
We estimated the fatal prevalence to be 14% of the total prevalence. The fatal prevalence is a major indicator of the demand for treatment for recurrences and palliation. Many of these cases are metastatic and can be cured in only in a minority of cases, although complete (but temporary) or partial remission can be obtained in most patients by modern treatments. It is likely that the interval between recurrence and death is increasing as palliative care becomes more effective. The fatal prevalence and demand for palliative treatment may also be increasing.
Other studies to estimate the health demand over time of colon cancer patients have been published. Three such studies, concerned with colon and rectal cancer combined, gave prevalence by Dukes' stage at diagnosis [1517
]. Specifically, they estimated the numbers of patients: (i) at Dukes' stage B or C at diagnosis who required intensive follow-up and probably also treatment for disease progression; (ii) with metastatic disease at diagnosis who required palliative treatment; and (iii) with Dukes' A who did no usually require further treatment after surgery (i.e. were cured), although endoscopy to check for metachronous tumors was generally prescribed. The results of these studies were closely concordant.
Two of the above-mentioned studies were population-based [16, 17
], and used prevalence estimates from French cancer registry areas to project estimates of French national prevalence rates. A special survey of cases assessed the risk of local recurrence and metastatic disease. This allowed the authors to estimate the cumulative recurrence rate at up to 5 years after diagnosis, which, applied to the prevalent cases, allowed estimation of the number of recurrences by 1-year interval up to 5 years. Up to 5 years, 19% of prevalent cases were estimated to recur, compared with 17% in our study [see Table 1
, total of last column up to 5 years as proportion of total prevalence up to 5 years (column 7)].
An American study integrated colorectal cancer prevalence data with Medicare data to estimate the numbers of patients requiring care [18]. This study identified care prevalence, defined as the prevalent cases under care, and non-cure prevalence cases, defined as those not cured of their disease. The findings indicated that several years after diagnosis, many patients were still receiving treatment for their cancer or its sequelae. For patients diagnosed up to 20 years before the prevalence date (1996), 62% were still receiving some kind of health care: for the majority (38%) this consisted of surveillance only, but for 20% this was treatment for recurrence or metastases. Although these data refer to elderly patients (Medicare program), they were the first attempt to provide information useful for public health planners, clinicians and drug companies.
One direction for future prevalence studies would be to use increasingly available information such as prescription data, hospital and outpatient admissions, and pharmaceutical company databases to provide more refined breakdowns of the care needs of patients with cancer diagnoses. In Europe, such information is available only in certain countries. For other countries, unique identifiers are not assigned to patients to facilitate integration of information [19], or private treatment is widespread and treatments are rarely documented in accessible databases.
In conclusion, we feel it useful to project the prevalence breakdowns estimated in this paper to the current situation. The EUROPREVAL project [8] estimated that in Europe in 1992, the overall prevalence of colon cancer was 176 per 100 000, corresponding to 660 000 people living with a diagnosis of this cancer. If we assume similar prevalence figures for today and the coming years and apply the proportions estimated in this study, we arrive at estimates of 568 000 cured cases and 92 000 fatal cases, with 66 000 cases in the initial phase of treatment after diagnosis, 59 000 recurrent cases and 172 000 disease-free cases requiring intensive clinical follow-up. Clearly, the accuracy of these figures depends on the several assumptions used to arrive at them; however, similar projections can be made at the level of individual cancer registry populations. In such situations these estimates can be extremely useful for those planning the allocation of future health-care resources at the local level.
![]() |
Acknowledgements |
---|
Members of the EUROPREVAL Working Group are as follows. Austria: W. Oberaigner (Cancer Registry of Tyrol). Denmark: H. Storm, G. Engholm (Danish Cancer Society Institute of Cancer Epidemiology). Estonia: T. Aareleid (Estonian Cancer Registry). Finland: T. Hakulinen (Finnish Cancer Registry). France: G. Hédelin, (Bas-Rhin Cancer Registry); H. Lefevre (Calvados Digestive Cancer Registry); J. Mace-Lesec'h (Calvados General Cancer Registry); J. Faivre (Côte d'Or Digestive Cancer Registry); G. Chaplain (Côte d'Or Gynaecologic Cancer Registry); P. M. Carli (Côte d'Or Malignant Haemopathies Registry); P. Arveux (Doubs Cancer Registry); J. Estève (University of Lyon); M. Colonna (Isère Cancer Registry); N. Raverdy, P. Jun (Somme Cancer Registry). Germany: J. Michaelis (German Registry of Childhood Malignancies); H. Ziegler, C. Stegmaier (Saarland Cancer Registry). Iceland: H. Tulinius (Icelandic Cancer Registry). Italy: R. Capocaccia (Project Leader); I. Corazziari, R. De Angelis, S. Francisci, S. Hartley, F. Valente, A. Verdecchia, A. Zappone (National Institute of Health, Rome); F. Berrino, G. Gatta, A. Micheli, E. Mugno, M. Sant (National Cancer Institute, Milan); P. Crosignani (Lombardy Cancer Registry); E. Conti, V. Ramazzotti (Latina Cancer Registry); M. Vercelli, C. Casella, A. Puppo (Liguria Cancer Registry, University of Genova); M. Federico (Modena Cancer Registry); M. Ponz De Leon (Modena Colorectal Cancer Registry); V. De Lisi (Parma Cancer Registry); R. Zanetti, S. Rosso (Piedmont Cancer Registry); C. Magnani (Piedmont Childhood Cancer Registry); L.Gafà, R. Tumino (Ragusa Cancer Registry); F. Falcini (Romagna Cancer Registry); E. Paci, E. Crocetti (Tuscany Cancer Registry); S. Guzzinati, P. Zambon (Venetian Cancer Registry). Poland: J. Rachtan (Cracow Cancer Registry); M. Zwierko, M. BielskaLasota (Warsaw Cancer Registry). Slovakia: I. Plesko (National Cancer Registry of Slovakia). Slovenia: V. PompeKirn (Cancer Registry of Slovenia). Spain: I. Izarzugaza (Basque Country Cancer Registry); A. Izquierdo (Girona Cancer Registry);I. Garau (Mallorca Cancer Registry); E. Ardanaz, C. Moreno (Navarra Cancer Registry); J. Galceran (Tarragona Cancer Registry); V. Moreno (Catalan Institute of Oncologia). Sweden: T. Möller, H. Anderson (Southern Swedish Regional Tumour Registry). Switzerland: J. Torhorst (Basel Cancer Registry); C. Bouchardy, J. M. Lutz, M. Usel (Geneva Cancer Registry); J. E. Dowd (WHO, Geneva). The Netherlands: J. W. W. Coebergh, M. Janssen-Heijnen (Eindhoven Cancer Registry); R. A. M. Damuhis (Rotterdam Cancer Registry). UK: R. Black, V. Harris, D. Stockton (Scottish Cancer Intelligence Unit); T. W. Davies (East Anglian Cancer Registry); M. P. Coleman, S. Harris (London School of Hygiene and Tropical Medicine); E. M. I. Williams (The Merseyside and Cheshire Cancer Registry); D. Forman, R. Iddenden (Northern and Yorkshire Cancer Registry and Information Service & Centre for Cancer Research); M. J. Quinn (Office for National Statistics); M. Roche (Oxford Cancer Intelligence Unit); J. Smith (South and West Cancer Intelligence Unit); H. Moller (Thames Cancer Registry); P. Silcocks (Trent Cancer Registry); G. Lawrence, K. Hemmings (West Midlands Cancer Intelligence Unit).
Received for publication November 7, 2003. Accepted for publication February 19, 2004.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2. Giles G. How important are estimates of cancer prevalence? Ann Oncol 2002; 13: 815816.
3. Crosignani P, Berrino F, Tagliabue G, et al. Italy, Varese province. (ed. Parkin D, Whelan J, Ferlay J, et al.), Cancer Incidence in Five Continents, Vol VII. IARC Scientific Publications No. 143. Lyon: IARC Press 1997; 566569.
4. Capocaccia R, Colonna M, Corazziari I et al. Measuring cancer prevalence in Europe: the EUROPREVAL project. Ann Oncol 2002; 13: 831839.
5. Verdecchia A, De Angelis R, Capocaccia R et al. The cure for colon cancer: results from the EUROCARE study. Int J Cancer 1998; 77: 322329.[CrossRef][ISI][Medline]
6. Krogh V, Micheli A. Measure of cancer prevalence with a computerized program: an example on larynx cancer. Tumori 1996; 82: 287290.[ISI][Medline]
7. Capocaccia R, De Angelis R. Estimating the completeness of prevalence based on cancer registry data. Stat Med 1997; 16: 425440.[CrossRef][ISI][Medline]
8. Micheli A, Mugno E, Krogh V et al. Cancer prevalence in European registry areas. Ann Oncol 2002; 13: 840865.
9. World Health Organization. International Classification of Diseases, 9th revision. Geneva: WHO 1997.
10. Tomatis L (ed.). Cancer: Causes, Occurrence and Control. IARC Scientific Publications No. 100. Lyon: IARC Press 1990.
11. Berrino F, Capocaccia R, Coleman M, et al. Survival of cancer patients in Europe: the EUROCARE-3 study. Ann Oncol 2003; 14 (Suppl 5): V1V155.[CrossRef][Medline]
12. Gatta G, Capocaccia R, Sant M, et al. Understanding variations in survival for colorectal cancer in Europe: a EUROCARE high resolution study. Gut 2000; 47: 533538.
13. Galandiuk S, Wieand HS, Moertel CG, et al. Patterns of recurrence after curative resection of carcinoma of the colon and rectum. Surg Gynecol Obstet 1992; 174: 2732.[ISI][Medline]
14. Rosselli DT, Palli D, Cariddi A et al. Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National Research Council Project on Breast Cancer Follow-up. JAMA 1994; 271: 15931597.[Abstract]
15. Gatta G, Francisci S, Ponz de Leon M. The prevalence of colorectal cancer in Italy. Tumori 1999; 85: 387390.[ISI][Medline]
16. Colonna M, Grosclaude P, Launoy G et al. Estimation of colorectal cancer prevalence in France. Eur J Cancer 2001; 37: 9396.[ISI][Medline]
17. Benhamiche-Bouvier AM, Clinard F, Phelip JM et al. Colorectal cancer prevalence in France. Eur J Cancer Prev 2000; 9: 303307.[CrossRef][ISI][Medline]
18. Mariotto A, Warren JL, Knopf KB, Feuer EJ. The prevalence of patients with colorectal carcinoma under care in the U.S. Cancer 2003; 98: 12531261.[CrossRef][ISI][Medline]
19. Storm H. Available data sources and linkage possibilities using cancer registry data in the evaluation of clinical cancer care. In Sankila R, Black R, Coebergh JW, et al. (eds): Evaluation of Clinical Care by Cancer Registries. IARC Technical Publications No. 37. Lyon: IARC Press 2003.