The likelihood of a breast cancer responding to an endocrine agent, whether in the primary cancer or in metastatic disease, is greater in those patients whose tumours have much higher levels of ER expression

The likelihood of a breast cancer responding to an endocrine agent, whether in the primary cancer or in metastatic disease, is greater in those patients whose tumours have much higher levels of ER expression. with trepidation that one even asks the question of whether clinical trials are still essential, since it would seem to challenge the work of many eminent physicians and statisticians. But what is surely open to challenge is whether in an era of much greater understanding of the biology of a disease, and with the advent of brokers that are only expected to be active in subsets of patients, the traditional inclusive, phase III design is still BAY 11-7085 the best model to provide a robust estimate of the effect of a novel BAY 11-7085 intervention in the relevant population of patients. Clinical trials are simply well-designed clinical experiments to test a particular hypothesis: and most anticipate a future in which the conversation between a treatment and the biology of the disease is pivotal to the hypothesis under scrutiny. Furthermore, many founder clinical trials have changed practise and outcomes for women with breast cancer. The analysis of improvements in the population of British Columbia breast cancer [1] or in French patients presenting with metastatic disease [2], as well as indirect comparisons between adjuvant chemotherapy trials (Barros C, personal communication), all confirm that a series of well-designed phase III trials in patient populations not selected for on the basis of tumour biology (and only sometimes on the basis of risk of relapse) have revolutionised the outcomes for women with breast cancer. It is evident, however, that none of our current armamentarium of systemic treatments works BAY 11-7085 in every patient with breast cancer. Even hormonal therapy, long since exhibited by the Oxford meta-analyses of many large trials as saving the lives of around 1 in 10 patients diagnosed with breast cancer, does not work in all patients. An intriguing thought experiment is therefore to consider how we might have re-designed those founder clinical trials with our current knowledge. Data published by Allred and colleagues demonstrate using immunohistochemistry that patients whose breast cancers have no, or very low BAY 11-7085 levels of, oestrogen receptor (ER) have a poorer outcome [3]. These data led to the US Food and Drug Administration definition of ER-positive breast cancer being anyone whose cancer has an Allred category score 3 (equivalent to at least 1% of cells staining moderately, or at least 10% of cells with any degree of staining). These data, together with Oxford overview data demonstrating no evidence of benefit for the use of adjuvant tamoxifen in women with cancers deemed to be ER-negative by a different, biochemical assay, have led some to believe that all patients with ER-positive breast cancers, as defined by the US Food and Drug Administration, should have adjuvant hormonal therapy. In contrast, there are many data confirming there to be subgroups of patients who have very endocrine-sensitive cancers, who in an age of targeted therapies would seem to be the ideal subpopulation in which one would test the benefit of adjuvant hormonal therapy. Clues to the diagnostic for this sensitive subgroup can be found in many studies. The likelihood of a breast cancer responding to an endocrine agent, whether in the primary cancer or in metastatic disease, is usually greater in those BAY 11-7085 patients whose tumours have much higher levels of ER expression. Furthermore, retrospective analyses of the Arimidex, Tamoxifen Alone or in Combination trial in fact suggested that those women whose cancers were ER-positive but progesterone receptor (PgR)-unfavorable had greater additional benefits from the use of 5 years’ anastrazole in place of tamoxifen, than their counterparts whose tumours had significant levels of PgR. Other analyses in that same study, however, as well as the Breast International Group 1-98 and Tamoxifen and Exemestane Adjuvant Multicentre trials, do not confirm this differential benefit – and rather suggest that ER-positive cancers with either coexpression of HER2 or lack of expression of PgR do worse with any endocrine agent, and that the relative level of additional benefit for the use of an aromatase inhibitor may be comparable across all types of ER-positive breast cancers. So how would wenowdesign an adjuvant trial with a control, no-hormonal therapy arm? If we target the trial at those patients with very hormone-sensitive disease (ER strongly positive, PgR-positive and/or HER2-unfavorable), we would enrich the population for those patients with better outcomes, greater relative IMMT antibody benefit from the therapy and longer time to recurrence. The trial would certainly be positive, but might take many years to conclude.