Breast cancer is the most common form of cancer in women, making up 16 percent of all cases of female cancer worldwide. Estrogen receptor (ER) positive breast cancers, whose growth is stimulated by the hormone estrogen, make up for three quarters of all breast cancer cases. Following successful treatment of ER positive breast cancer, some women will undergo relapse, even after five years of adjuvant hormone therapy. The mechanistic reasons underlying relapse are not wholly understood although markers for proliferation and estrogen receptor expression are used to help predict the chances of recurrence. Giampaolo Bianchini from the San Raffaele Scientific Institute, Italy, and colleagues investigated whether these markers could be used to better identify ER positive breast cancer patients at risk of early versus late relapse. Bianchini explains more about their key results, published in a recent study in Breast Cancer Research, and why these differences in relapse may occur.
Following your medical training, what led to your interest in oncology and specifically breast cancer research?
Actually, my interest in oncology started before I began my medical training. Unfortunately, my best friend died of cancer and at that time I decided that I would become a medical oncologist. The reason why I specifically dedicated myself to breast cancer was more accidental. During my residency I worked in the division which was previously directed by Gianni Bonadonna and at that time was led by Luca Gianni. They contributed so much to the advancement of treatments for breast cancer, that I decided to follow their path.
Can you tell us a bit about the prevalence of relapse in estrogen receptor positive breast cancer and current approaches to preventing this?
Estrogen receptor positive breast cancers are extremely heterogeneous molecular diseases with different clinical behaviors, from indolent tumors at low risk of relapse to very aggressive tumors with a high risk of recurrence, similar to triple negative breast cancers. For reducing the risk of recurrence, the therapeutic options available in early breast cancer include hormonal treatment with or without additional chemotherapy. The conventional duration of endocrine treatment is five years. A relevant characteristic of these tumors is the almost continuous and long-lasting risk of relapse over more than one decade despite the treatment received. Thus, an additional treatment option in patients receiving five years of adjuvant tamoxifen is extending the endocrine treatment with letrozole for post-menopausal patients, or even with tamoxifen. Both these therapeutic approaches have been shown to reduce the risk of recurrence and death. However, the treatment benefit from extended endocrine therapy has to be weighed against short and long term side effects. This prompted the urgent need to identify which patients will actually benefit the most from this treatment.
What was the aim of your study and what were your key findings?
Proliferation markers and estrogene-related genes have been shown to be relevant for predicting the risk of relapse in estrogene receptor positive tumors treated with endocrine therapy. However, if their prognostic value is different at different time points was not well characterized. In our work, we examined, in a time-dependent manner, whether the combination of proliferation markers and estrogen receptor-related gene expression could improve the ability of these variables to predict the risk of recurrence in the first five years (early relapses) and from five to ten years (late relapses), in women with ER-positive breast cancer treated with five years of tamoxifen. We found a significant time-varying effect for different biomarkers, which means that the molecular characteristics and the corresponding biological processes associated with early and late relapses are significantly different. Early relapses, especially in the first 2.5 years, are highest in highly proliferative tumors with low expression of estrogen-related genes. This group appeared to be strongly enriched in tumors intrinsically resistant to both tamoxifen and aromatase inhibitors, as also suggested by the poor response to neoadjuvant letrozole in this group. In the late period, after five years of adjuvant tamoxifen, both the group with highly proliferative tumors with high expression of estrogen-related genes and the group with low proliferation tumors with low expression of estrogen-related genes were at high risk of relapse.
What do you think underlies the differences in early versus late relapse between patients with highly proliferative/low estrogen-related gene expression cancers and highly proliferative/high estrogen-related gene expression cancers, respectively?
High proliferation ER-positive tumors are at higher risk of relapse in both untreated and tamoxifen-treated ER-positive breast cancer. However, in patients treated with adjuvant tamoxifen, the prognostic value of the estrogen-related genes in this group with high proliferation tumors had a very strong time-dependent effect. In the early period, especially within the first 2.5 years, the group with high expression of estrogen-related genes had a very low risk of relapse, likely reflecting the higher likelihood of benefit from tamoxifen in this group. Conversely, the same group in the late period showed a significantly higher risk of relapse compared to the group with low expression of estrogen-related genes. Given that we did not observe a similar behaviour of these markers in untreated breast cancer, we can speculate that some tumors with high proliferation and high expression of estrogen-related genes could have relapsed earlier without treatment, but relapse was delayed due to adjuvant therapy. These tumors are likely to retain at least a partial endocrine sensitivity and could be the best candidates for extended endocrine therapy especially with an aromatase inhibitor.
How do you think your results could impact risk assessment for relapse in patients with estrogen-sensitive breast cancer?
Different multigene assays have been developed with the purpose of identifying those ER-positive patients treated with endocrine therapy that have a risk of relapse low enough to be safely spared chemotherapy. One of these assays, the Oncotype DX, is already broadly used in clinical practice for this purpose. Our approach for risk assessment should be considered in combination with these existing multigene assays to refine risk prediction in the late relapse period and to improve clinical decision making about extended endocrine treatment.
What further studies are needed to translate your findings in to the clinic? What’s next for your research?
Even though in our study we confirmed our findings in two cohorts of patients, independent validation – hopefully performed with a prospective-retrospective approach – is needed. This means using data generated in a prospective clinical trial in a retrospective way. The second step needed to translate our findings in to the clinic is to define an assay which could be performed using formalin-fixed paraffin-embedded tissue in clinical practice and to identify relevant cut-offs for this assay to assist clinical decision making. Our group is evaluating a better strategy to accomplish both of these steps.