Trends in breast cancer incidence, survival, and mortality

Abstract

Author's replySir—Breast cancer death rates had been rising slowly for decades in the UK, as in many other European countries, until the 1980s. Around this time, treatments of proven efficacy began to be used widely and the breast cancer death rate in middle age suddenly began to fall, and it is now about 30% lower than it was 10 years ago. The key difference between the fall in stomach cancer deaths, which has been in progress for several decades, and that in breast cancer deaths, which is of recent origin, lies in the suddenness of this change between slow rise and rapid fall, which is far too sharp for the decrease to be plausibly ascribable to real changes in incidence. Although trends in breast cancer mortality in old age may be subject to substantial artefacts, those in middle age are not.1Doll R Peto R The causes of cancer.J Natl Cancer Inst. 1981; 66: 1193-1308Google Scholar, 2Peto R Trends in US cancer onset rates.in: Peto R Schneiderman M Quantification of occupational cancer. Banbury Report 9. Cold Spring Harbor Laboratory, 1981: 269-284Google Scholar and these show a substantial recent benefit. In contrast, sudden increases in the recorded incidence rates have occurred in recent years that do not reflect real changes in incidence, and that artificially improve the 5-year survival rates. The graphs provided by Michel Coleman and colleagues illustrate how large the artefactual increases in the recorded breast cancer incidence (and hence survival) rates in the UK have been in recent years. The use of breast cancer survival rates to compare the quality of health care in different populations or at different times is a practice that we were criticising, and the continuation of this practice is not evidence that our criticisms of it are unjustified.One major contributor to the recent reduction in breast cancer mortality has been the widespread use of hormonal treatment, particularly tamoxifen, for patients with breast cancer whose primary tumour is oestrogen-receptor positive; whether such treatments are also of some net value for some women with oestrogen-receptor-negative disease is, as J R Benson and A Purushotham suggest, an important open question. The stress on tamoxifen in our press material (which was written by me, not by ICRF) appropriately reflects the strength of the randomised evidence that such treatment improves long-term survival;3Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials.Google Scholar this is not true of the psychosocial interventions and holistic approach advocated by Heather Goodare, nor is it true of the tamoxifen analogue discussed by Henri Pujol. Author's reply Sir—Breast cancer death rates had been rising slowly for decades in the UK, as in many other European countries, until the 1980s. Around this time, treatments of proven efficacy began to be used widely and the breast cancer death rate in middle age suddenly began to fall, and it is now about 30% lower than it was 10 years ago. The key difference between the fall in stomach cancer deaths, which has been in progress for several decades, and that in breast cancer deaths, which is of recent origin, lies in the suddenness of this change between slow rise and rapid fall, which is far too sharp for the decrease to be plausibly ascribable to real changes in incidence. Although trends in breast cancer mortality in old age may be subject to substantial artefacts, those in middle age are not.1Doll R Peto R The causes of cancer.J Natl Cancer Inst. 1981; 66: 1193-1308Google Scholar, 2Peto R Trends in US cancer onset rates.in: Peto R Schneiderman M Quantification of occupational cancer. Banbury Report 9. Cold Spring Harbor Laboratory, 1981: 269-284Google Scholar and these show a substantial recent benefit. In contrast, sudden increases in the recorded incidence rates have occurred in recent years that do not reflect real changes in incidence, and that artificially improve the 5-year survival rates. The graphs provided by Michel Coleman and colleagues illustrate how large the artefactual increases in the recorded breast cancer incidence (and hence survival) rates in the UK have been in recent years. The use of breast cancer survival rates to compare the quality of health care in different populations or at different times is a practice that we were criticising, and the continuation of this practice is not evidence that our criticisms of it are unjustified. One major contributor to the recent reduction in breast cancer mortality has been the widespread use of hormonal treatment, particularly tamoxifen, for patients with breast cancer whose primary tumour is oestrogen-receptor positive; whether such treatments are also of some net value for some women with oestrogen-receptor-negative disease is, as J R Benson and A Purushotham suggest, an important open question. The stress on tamoxifen in our press material (which was written by me, not by ICRF) appropriately reflects the strength of the randomised evidence that such treatment improves long-term survival;3Early Breast Cancer Trialists' Collaborative Group. Tamoxifen for early breast cancer: an overview of the randomised trials.Google Scholar this is not true of the psychosocial interventions and holistic approach advocated by Heather Goodare, nor is it true of the tamoxifen analogue discussed by Henri Pujol. Trends in breast cancer incidence, survival, and mortalityThe continuing fall in breast cancer mortality in England and Wales reported by Richard Peto and colleagues (May 20, p 1822)1 is extremely welcome. We agree that improvements in the management of breast cancer since the 1970s have prevented large numbers of breast cancer deaths in middle age. This is reflected in the mortality trends, but it is also clear from joint examination of trends in incidence and mortality2 and survival.3,4 The dismissal by Peto and colleagues of cancer incidence and survival rates for temporal and geographic comparisons is surprising. Full-Text PDF Trends in breast cancer incidence, survival, and mortalityThe report from Richard Peto and colleagues1 of a 25–30% reduction in breast cancer mortality over the past decade is encouraging and consistent with data from meta-analyses of adjuvant systemic therapies. The widespread incorporation of tamoxifen into standard clinical practice during the late 1980s and early 1990s has figured substantially in these improved mortality statistics, though the relative contributions of tamoxifen and other factors, such as chemotherapy and screening, remain to be quantified. Full-Text PDF Trends in breast cancer incidence, survival, and mortalityThe way that Richard Peto and colleagues' letter,1 (NB, correspondence, not research letter, nor study) about the welcome recent decline in breast-cancer mortality was trumpeted in the media was most extraordinary.2–5 Peto himself was heard to say on the radio that the decline was largely a result of widespread tamoxifen use. However, the letter says: “this substantial reduction in national mortality rates has come not from a single research breakthrough, but from the careful evaluation and adoption of many interventions, each responsible on its own for only a moderate reduction in breast-cancer mortality.” Full-Text PDF Trends in breast cancer incidence, survival, and mortalityBreast cancer mortality remains a major challenge to public health, even though Richard Peto and colleagues report a consistent decrease.1 Screening programmes and adjuvant treatments have been developed over the past decade and edge us towards the goal of reduced mortality. However, breast cancer mortality remains a dominating concern for the oncologist. Therefore, chemoprevention with tamoxifen has been investigated as a third way of decreasing the risk of death from breast cancer. Three studies aimed at assessing the magnitude of tamoxifen chemoprevention have produced controversial results. Full-Text PDF Trends in breast cancer incidence, survival, and mortalityThe letter by Richard Peto and colleagues1 made headlines throughout the world. They argued that there has been a significant decline in breast cancer mortality in the UK and the USA. In the popular press coverage, Peto was quoted as attributing the decline to greater use of tamoxifen. It is unusual that a letter to the Editor achieved such significant recognition. The correspondence did not directly assess tamoxifen nor were the methodologies for the study described. Perhaps more challenging is that the data offered by Peto seemed to be in conflict with other published reports. Full-Text PDF