For patients with triple negative breast cancer who have a minimal response to neoadjuvant chemotherapy, do you do additional testing to sub-categorize the cancer and find a potential target (e.g. androgen receptor, etc)?
We often encounter this scenario when we have residual disease after Neo-adjuvant chemotherapy in TNBC and I think about these 3 options:
- ECOG Trial/ NCT02445391: Phase III Trial randomizing Platinum (4 cycles of cisplatin or Carboplatin) versus oral Capecitabine x 6 cycles. (Requires residual disea...
There are a couple of options for a patient who does not respond to neoadjuvant chemotherapy and has residual disease. The first would be to consider offering the patient to participate in SWOG 1418 that randomizes women to pembrolizumab vs. usual care. There is some evidence that immunotherapy may ...
If I'm treating off-study, I would consider 8 cycles of capecitabine on the basis of the CREATE-X data. Treatment improved 5-year DFS by 8 percentage points with manageable toxicity. We need more time for OS data and reimbursement for early-stage disease.
While doing this would be very intriguing, currently there is no standard approved agents for sub-groups if triple negative breast cancer, unless clinical trials are considered.
Although it may be helpful to know the AR status, there is no standard treatment based on the additional subcategories outside of clinical trials. For someone who had minimal response to chemotherapy, I would explore adjuvant immunotherapy trials (some of which allows for concurrent radiation). The ...
Enrollment in a clinical trial, if possible, is the best option. ASCENT-05 is a trial that hopes to address this question. The trial randomizes TNBC patients with residual disease after neoadjuvant chemotherapy and surgery to either Pembrolizumab plus Sacituzumab OR Pembrolizumab +/- Capecitabine.
In view of the absence of clinical trial evidence that genomic data can be used to guide adjuvant systemic therapy for women with non-metastatic triple negative breast cancer, I don't know how I could use that data. I, therefore, typically don't obtain it.