Would you consider adding pembrolizumab to adjuvant chemotherapy in a patient with stage 3 TNBC who did not receive neoadjuvant treatment?

It seems that we might be missing some key clinical information here.

However, I assume that this was high risk enough - given the ki-67, potentially LN positive, and larger size tumor to make the treating physician concerned.

As a first point, while the ctDNA reflects (based on accumulating data) potential minimal residual disease vs micrometastatic disease, suggesting that she is likely needing therapy. However, many of these ctDNA studies have been done, after the patient has undergone systemic therapy - supporting the resistant clone. In this specific case scenario, conventional chemotherapy may be sufficient to induce eradication of the MRD or micromets. Therefore, at least standard care adjuvant therapy including ddAC and taxol would need to be considered. 

If the patient had a very high-risk disease aside from the ctDNA - like T stage, N stage, etc, the question here is whether to add either carboplatin or carboplatin + Immune checkpoint inhibitor. If we use the same logic that we have been using in HER2 positive - if the regimen has shown higher induction of the response in the neoadjuvant setting, why not applying the same in the adjuvant setting. And of course, accumulating data also suggests that postpartum cancer or mid-pregnancy cancer tend to be more aggressive in general (potentially through an immune-mediated mechanism, the enhanced growth pattern of the normal/cancer cells). 

Given all of these, if the patient has TNBC with a very high risk of progression/recurrence, it might be worth of shot - at least go over this as a potential option with the patient, and submit it to the insurance and see if this will be approved. 

Based on the current data available, I would not recommend administering pembrolizumab. There is definitely some detailed clinical data missing. I speculate that we are talking about no BRCA mutation and TNBC setting. At this point, there is no clinical trial data to support the use of adjuvant pembrolizumab for the patient. It is important to note that the use of adjuvant pembrolizumab should only be considered after the patient has received neoadjuvant immunochemotherapy, and it should be clearly separated from the standard of care. You can not justify the toxicity when the efficacy is not established.

It would be typical for high risk patients to have had neoadjuvant chemotherapy. The KEYNOTE-522 study used pembro as paclitaxel/carbo-pembro/AC neoadjuvantly, and continued pembro alone for 9 cycles in the adjuvant setting. So high risk TNBC patients should ideally get neoadjuvant pembro/chemo followed by adjuvant pembro. How to combine this with adjuvant capecitabine is another story.

For high risk patients who, for some reason, did not have neoadjuvant chemotherapy or did receive it without pembro, we have no data on adjuvant pembro in this case. 

If she's BRCA positive, I'd favor olaparib per OlympiA trial over capecitabine per CREATE-X. The answer to the posed question, however, will probably wait until we have the results from SWOG S1418 (completed enrollment early this summer).