Journal of the National Cancer Institute Advance Access originally published online on July 29, 2008
JNCI Journal of the National Cancer Institute 2008 100(15):1117; doi:10.1093/jnci/djn217
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2008. Published by Oxford University Press.
CORRESPONDENCE |
Re: Should Observational Studies Be a Thing of the Past?
Affiliations of author: Department of Epidemiology, University of Washington, Seattle, WA, and the Fred Hutchinson Cancer Research Center, Seattle, WA
Correspondence to: Noel S. Weiss, MD, DrPH, Department of Epidemiology, University of Washington, Box 357236, Seattle, WA 98195 (e-mail: nweiss{at}u.washington.edu).
In her editorial, Kathleen Pritchard (1) stated that the results of several randomized trials "clearly provide a definite answer" regarding the influence of hormone replacement therapy (HRT) administered after a diagnosis of breast cancer on the prognosis of women with this disease. The trials to which she referred are the multinational Hormonal Replacement Therapy After Breast Cancer—Is It Safe? (HABITS) study (2,3) and the Stockholm Randomized Trial (4). The "definite answer" appeared to have been provided by the observation in the HABITS trial of a substantial increase in the incidence of local recurrence and contralateral cancer in the women assigned to HRT—there were 29 such events in the HRT arm, compared with only 9 among women in the equal-sized control arm. However, no such association was seen in the Stockholm study—the corresponding numbers were 8 and 8. Chance conceivably could account for difference in results between the studies, but a formal test of statistical heterogeneity suggests otherwise—the P value was .02. Another possible explanation relates to the difference in the type of progestogen typically given to women in the two studies—norethisterone acetate in the HABITS trial vs medroxyprogesterone acetate in the Stockholm trial—but in the HABITS trial, use of estrogen alone was associated with a recurrence risk similar to that of use of combined hormone therapy (3). Finally, even though in both studies an effort was made to standardize the frequency of follow-up of study participants, the lack of blinding possibly could have led to differentially complete ascertainment of the more subtle forms of recurrence between women in the two arms of the trials.
The lack of blinding in the trials argues that relatively more weight should be given to the results for the relatively more serious outcomes of breast cancer mortality and the development of metastatic disease. Examination of these outcomes does not suggest an adverse influence of HRT. There was no increase in breast cancer mortality in either trial. In the HABITS trial, during an average of 4 years of follow-up there were six deaths from breast cancer among 221 women assigned to receive HRT, vs five deaths among an equal number assigned to not receive HRT (3). In the Stockholm trial (4), there were two deaths from breast cancer among the 188 women assigned to receive HRT during an average of 4.1 years of follow-up, vs four such deaths among 190 women in the control arm. In the HABITS trial (3), distant metastases developed in 10 women assigned to HRT and in eight not so assigned; the corresponding numbers in the Stockholm study were three and five. Thus, neither trial observed an appreciable increase in risk of metastatic disease associated with administration of HRT.
I wholeheartedly agree with Dr Pritchard that "interventional studies with robust controlled designs" should be implemented when feasible. However, in the case of HRT administration following a diagnosis of breast cancer, I believe that in spite of several randomized trials that sought to address the question there is not yet convincing evidence that the earlier results of nonrandomized studies, which observed no altered risk of recurrence or mortality associated with hormone use, were invalid.
REFERENCES
1. Pritchard KI. Should observational studies be a thing of the past? J Natl Cancer Inst. (2008) 100(7):451–452.
2. Holmberg L, Anderson H. HABITS (hormonal replacement therapy after breast cancer—is it safe?), a randomized comparison: trial stopped. Lancet (2004) 363(9407):453–455.[CrossRef][Web of Science][Medline]
3. Holmberg L, Iversen OE, Rudenstam CM, et al. on behalf of the HABITS Study Group. Increased risk of recurrence after hormone replacement therapy in breast cancer survivors. J Natl Cancer Inst. (2008) 100(7):475–482.
4. von Schoulz E, Rutqvist LE. on behalf of the Stockholm Breast Cancer Study Group. Menopausal hormone therapy after breast cancer: the Stockholm Randomized Trial. J Natl Cancer Inst. (2005) 97(7):533–535.
Response to this Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 1117-1118.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||