Editor's Choice | This Week in BMJ | Press releases

BMJ No 7119 Volume 315 Letters Saturday 22 November 1997 New method for expressing survival in cancer Editor, Vaidya and Mittra, in their paper, reported an alternative way of presenting survival data in cancer patients.(1) This, they hope will be more acceptable to patients. There are, however, some flaws with the paper. Age, apart from the usual variables used in conventional survival curves, is a new variable in this system of normal remaining life (NRL) fraction. Most cancer incidences, including breast,(2) rises with age. By the virtue of increasing age specific incidence and the relative stasis of normal life expectancy, once adulthood is reached, the NRL fraction will be nearer one with increasing age. Therefore, the data used to generate NRL fraction curve, without correction for non-normally distributed age incidence or being specific for various age groups, will lead to false assurances in the younger age groups. The impression given in table 1 is that, in node negative disease, 81% of forty-year olds will live for 16 years while only 68% of 60-year olds will survive for 15 years. Similarly, while 20% of 40-year olds with 4-node involvement live for 16 years, no 60-year old will live that long. In the conventional method, 48% of patients with 1-4 nodal involvement will survive for 10 years, whereas in the new method, only 38% of 60-year olds will survive for three-quarters of this period.                         Table 1 - Comparison of survival estimates by conventional and real life expectancy Conventional method Real life expectancy method Node negative 88% Survive 5 years 89% Survive a fifth of normal remaining life (6.4 years at age 40, 3 years at age 60) 82% Survive 10 years 81% Survive half normal remaining life (16 years at 40, 7.5 years at 60) 68% Survive full normal remaining life (cure) (32 years at 40, 15 years at 60) 1-4 Nodes positive 66% Survive 5 years 65% Survive a fifth normal remaining life (6.4 years at 40, 3 years at 60) 48% Survive 10 years 38% Survive half normal remaining life (16 years at 40, 7.5 years at 60) 28% Survive full normal remaining life (cure) (32 years at 40, 15 years at 60) >4 Nodes positive 40% Survive 5 years 38% Survive a fifth normal remaining life (6.4 years at 40, 3 years at 60) 22% Survive 10 years 20% Survive a half normal remaining life (16 years at 40, 7.5 years at 60) None survive full normal remaining life (cure) "Living all normal remaining life is equivalent to cure," as the authors assert, is blatantly untrue. A cure denotes freedom from recurrences or relapses and metastasis, and not just death from the disease. The NRL fraction curve shows that, in some patients with none or with 1-4 nodes, mortality from the disease occurred after the "full life span"! Is it their assertion also, that 68% of all node-negative patients are cured!! A new way of expressing cancer mortality is essential but the NRL fraction curve falls short of this. The number of years lived by any age group can be calculated from conventional survival curve. The initial expression of survival as fraction of NRL and the re-conversion back to the number of years expected to live distorts the original data. The idea of expressing survival in terms of normal remaining life has a great potential in determining appropriateness of medical management, but it needs a more rigorous way of calculating it. O A Adedeji Department of General Surgery, Wansbeck General Hospital, Ashington, Northumberland NE63 0SA Reference 1 Vaidya J S, Mittra I. Fraction of normal remaining life span: a new method for expressing survival in cancer. BMJ 1997; 314:1682-4. (7 June.) 2 McPherson K, Steel C M, Dixon J M. ABC of breast diseases. Breast cancer epidemiology, risk factors and genetics. BMJ 1994;309;1003-1006. Home | Current issue | Past issues | Classified ads | Career Focus | Feedback Collections | About this site | About the BMJ | BMA | Medline