Later-Line Choices for Patients With Metastatic TNBC

Selecting Frontline Therapy for TNBC After Progression on Adjuvant Therapy

During a Targeted Oncology case-based roundtable event, Elizabeth Ann Mittendorf, MD, discussed with participants their choice of treatment after progression for a patient with triple-negative breast cancer treated with adjuvant dose-dense doxorubicin plus cyclophosphamide with paclitaxel. This is the first of 2 articles based on this event.


A 48-year-old woman with T1N1 triple-negative breast cancer (TNBC) received adjuvant dose-dense doxorubicin plus cyclophosphamide with paclitaxel (ddAC-paclitaxel), which she tolerated well. Eight months after completion of adjuvant therapy, she reported worsening fatigue. ​Laboratory results showed 1.5 times elevated levels of alanine transaminase (ALT) and aspartate transaminase (AST) above the upper limit of normal, and a CT scan showed 1 liver lesion and 2 left lung lesions. A biopsy of the liver lesion confirmed recurrent metastatic TNBC. A brain MRI was negative for brain metastases.

Genetic panel testing was negative for all detectable mutations. PD-L1 expression on immune cells was 0%. The patient was otherwise healthy with no significant comorbidities and had an ECOG performance score of 0.

What frontline therapy would you recommend for this patient with metastatic recurrence 8 months after adjuvant ddAC-paclitaxel for her TNBC?

What frontline therapy would you recommend for this patient with metastatic recurrence 24 months after adjuvant ddAC-paclitaxel for her TNBC?


  • Would your choice of treatment differ for progression after 24 months versus 8 months? Why?

SETH WANDER, MD: Does she has liver dysfunction from her disease?

ELIZABETH ANN MITTENDORF, MD, PHD: Yes, her AST and ALT were 1.5 times the upper limit of normal.

WANDER: I think that’s the more challenging aspect, to some extent, rather than the duration. Does she have end organ dysfunction, and some degree of impending visceral crisis? In my mind, that is driving more than the elapsed time in terms of how we would want to think about this.

I think it’s hard in practice; it’s easy to say, “If somebody has visceral crisis, we’re going to use aggressive combination chemotherapy.” It’s sometimes more challenging to define what exactly that means. So, if a patient is asymptomatic or minimally symptomatic, and has some mild transaminase elevation or something like that, is that enough to trigger you to think about treating more aggressively? It’s easier if the patient is having a lot of symptoms, or in more distress; but I think to some degree you worry about a patient who already has some organ dysfunction here, if you miss your window and you can’t debulk the disease fairly quickly, that you’re going to run into some serious problems pretty soon.

That, to me, in this case, is more important than 8 versus 24 months.I don’t know if anyone else feels strongly about the degree of liver elevation here as not being so worrisome because it’s relatively mild elevation; I think that you could argue that point, as well.

MITTENDORF: [Based on] 1.5 times upper limit of normal, I did not myself classify this patient as having severe liver dysfunction. But your point is well taken. Let’s say it’s not severe liver dysfunction, and she seems to otherwise be doing OK. Is there anybody who, based on the difference in the disease-free interval, would go back to a taxane for the 24 months?

HEATHER BENJAMIN, MD: I would be very tempted to do so if she didn’t have any significant neuropathy, and we could get a long duration. It could be something we could continue for a prolonged period of time.

WANDER: I agree. In that context, if you’re not worried about the liver, then it would make more sense with the duration to maybe go back to something that was well tolerated and effective previously.

PAUL UNGER, MD: We don’t know if it was effective previously, because this was given in the adjuvant setting.We don’t know if she responded to paclitaxel or not. This is someone who’s going to die of metastatic breast cancer. Some patients tolerate oral capecitabine and respond for what seems to be low-burden disease, in this case, and especially if it’s at 24 months. I think this is a marathon, not a sprint. Certainly, for someone with a crisis where you need a rapid response, that might not be the best agent, but we don’t know if she responded to any of these other things. I would try to think of this as what could potentially maintain her disease for a longer haul with better quality of life. I think taxane will catch up with patients for neuropathy, and it’s always there later.

SHEILA DONNELLY, MD: I agree with you. I would give capecitabine for 24 months, and I’d reserve a taxane for the next drug. I think you usually maintain quality of life better.

ANKUR MEHTA, MD: The carboplatin and gemcitabine combination is somewhat better than a single-agent taxane in this situation. I’m assuming this patient has not seen a carboplatin for TNBC.…And then, as a taxane, I would use eribulin [Halaven] in preference to paclitaxel.

PRABH SINGH, MD: Was the PD-L1 testing done on the liver disease, or on the original? They should repeat that, right?

MITTENDORF: Yes, it didn’t specifically say. But as we commented on before, liver is the least likely metastatic site to express PD-L1. So, let’s give them the benefit of the doubt, and suggest that they went back and looked at the primary tumor. I think we have to accept that it was PD-L1 negative.

SINGH: One question I am just now debating in my mind is, if we were using pembrolizumab [Keytruda] in the neoadjuvant setting, and then followed by adjuvant therapy, where does its role stand in the metastatic disease setting? That will be a question to be decided. How will we use IO [immuno-oncology] in patients with TNBC who have already seen IO in the neoadjuvant and adjuvant setting?

MITTENDORF: Yes, it’s a very interesting question. You don’t need to assess PD-L1 on the tumor to administer it in the preoperative neoadjuvant setting. So, I can’t say that I’ve seen that come up yet, but we don’t know about giving IO after IO. I think, in our practice at Dana-Farber Cancer Institute, we would likely—if a patient had gotten preoperative pembrolizumab, and then they recurred—send the tumor for PD-L1 testing, to at least think about that. But I’m confident that would be a tumor board discussion.