ADCs in the Frontline: Evolving Approaches in Metastatic TNBC

In a CancerNetwork® -hosted Satellite Session, a panel of Mayo Clinic oncologists discussed how recent frontline antibody-drug conjugate (ADC) data are reshaping treatment selection and sequencing in metastatic triple-negative breast cancer (TNBC), along with practical strategies for managing the toxicities that distinguish these agents.

The session was moderated by Roberto Leon-Ferre, MD, a medical oncologist at the Mayo Clinic Comprehensive Cancer Center in Rochester, Minnesota. He was joined by Karen S. Anderson, MD, PhD, an adjunct professor and medical oncologist at Mayo Clinic in Phoenix, Arizona; Eyad Al-Hattabb, MD, chair of oncology for the Mayo Clinic Health System in Eau Claire, Wisconsin; Shakeela Bahadur, MD, a breast medical oncologist at Mayo Clinic in Phoenix, Arizona; Felipe Batalini, MD, an assistant professor of oncology in the Mayo Clinic College of Medicine Mayo Clinic in Phoenix, Arizona; Saranya Chumsri, MD, an oncologist at Mayo Clinic in Jacksonville, Florida; Brenda J. Ernst, MD, a hematologist and oncologist at Mayo Clinic in Phoenix, Arizona; and Yanyan Lou, MD, a medical oncologist at Mayo Clinic in Jacksonville, Florida.

Tolerability and the Biomarker-Timing Dilemma in Frontline Treatment Selection

Leon-Ferre: Does anyone consider PARP inhibitors for any patients in the first line setting if they are also PD-L1–positive?

Batalini: I would say consider is a fair verb. Default is exactly what the guidelines state, but it’s not uncommon that patients want to do a non-chemotherapy approach. Realistically, we are getting the germline results a few weeks after we start first-line therapy, so it doesn’t become as much of a clinical conundrum as if we had the results up front…. The default for patients with PD-L1–positive disease is to combine pembrolizumab [Keytruda] with sacituzumab govitecan [Trodelvy] or chemotherapy.

Al-Hattab: I can tell you from a community perspective, although these are oral drugs, they are extremely hard to tolerate for most patients, at least that’s what we see with olaparib [Lynparza]. Most [patients] require some dose reduction to be able to tolerate it in our experience. Most of these patients have intra-auricular metastases, so you want a good, quick response. We do the chemotherapy with pembrolizumab as first-line treatment if they’re PD-L1 positive.

Leon-Ferre: What are the main toxicities that you have seen?

Al-Hattab: Extreme exhaustion and fatigue…. We had patients discontinue treatment when we used those because that adverse effect is the most pronounced. It’s not hematological. It’s mostly that feeling of exhaustion and fatigue.

Leon-Ferre: I certainly have had patients who had significant fatigue, but I wouldn’t say that I had a large number of them discontinue, or get dose reductions from that sense. When I’ve had that, they don't feel much better with an ADC. It's a plus or minus. Have others experienced non-anemia-associated fatigue?

Anderson: There are significant issues with sacituzumab govitecan and fatigue as well, so that may be an issue with any of them. I’ve probably seen more of it with sacituzumab govitecan than with the PARP inhibitors. I also wanted to follow something that Dr. Batalini pointed out: that sometimes the choices and strategies in the early line settings are dictated, somewhat, by getting the correlative biomarkers in a timely fashion, like getting the PD-L1 testing scores in time. Oftentimes we’re making quick treatment decisions for these patients, especially in the setting of rising liver function tests or worsening pulmonary function for a rapidly progressive disease. With sacituzumab govitecan, we can add in pembrolizumab if the PD-L1 staining is high. We may not have that yet in a very practical way, so we weigh this sort of back and forth amongst ourselves in which one we're going to start. There are different toxicity profiles with these, but you don’t always have the correlative data to leap in one way or another on this.

Evaluating PFS2 As an End Point

Leon-Ferre: What does everyone think about PFS2 as an end point? Is it meaningful?

Batalini: It’s meaningful. These are very different questions, because when wanting the OS benefit, it’s critical for new drugs asking the question whether they should be approved or not. The PFS2 question is more informative on not the value of the drug, but the value of the exact sequencing. Naturally, I expect that when you allow crossover, and mostly because this was a drug that was already approved in the second-line setting, it’s going to hurt the overall survival benefit. That’s consistent in many other trials, which speaks to the fact that perfect sequencing is not necessarily required. Using the drug in the second or later lines of therapy will likely salvage the majority of patients, but requiring OS is very critical for these new drugs that are much more expensive than the alternatives. When they get approved, the question becomes whether they should be moved to the first line as an option, not as a requirement. We need to be a little more flexible.

Anderson: One of the interesting questions is how these drugs might condition tumors for subsequent therapies in a specific or a non-specific way. The sense is that PFS2 ends up being better sometimes than we would expect, and that may be because we’re able to do crossovers for these. It also may be that these medications, in different ways, are conditioning this. We’ve learned this from immunotherapies that you have to look at the long haul, that you need to look at the overall survival, and even with PFS2, and we were seeing this some in the early days of immunotherapy. The second-line therapies were looking way better than we expected.

Treating Immunotherapy-Eligible Patients with Sacituzumab Govitecan

Leon-Ferre: If you have a patient who is eligible for immunotherapy, is there any patient scenario where you wouldn’t offer combining sacituzumab govitecan with pembrolizumab? Or would you favor monotherapy?

Al-Hattab: We’re always going to think about immunotherapy in patients with PD-L1–positive disease because, just like the abscopal effect sometimes we have with ADCs and immunotherapy, most of us think about it this way. I cannot think of a scenario where we do not do immunotherapy with sacituzumab govitecan.

Leon-Ferre: It’s hard to know if the ADCs may sensitize or revamp immune sensitivity. That's at least the hypothesis. Let's say the patient is just finishing up the adjuvant pembrolizumab portion and presents with metastatic disease with no disease-free interval. I don't know if you’ve had that experience, but I see patients who haven’t even finished their adjuvant pembrolizumab, and they’re already having onset of metastatic disease. Would you feel the same about continuing pembrolizumab in those patients?

Ernst: It’s very challenging. You don’t know the input of each individual agent.

Al-Hattab: I agree. I would be asking for opinions here.

Anderson: I'd probably be less enthusiastic about continuing the pembrolizumab in those patients, given that they're on it….

One question that comes up a lot in the community is whether there are alternative dosing opportunities with sacituzumab govitecan. With the 2 weeks on, 1 week off version, [the question is: is it challenging for people, especially over the longer haul? The maintenance of blood counts especially, and the need for supplemental G-CSF and others. It ends up being quite a challenge to get these medications in, even at dose-reduced levels, especially in these patients who are heavily pretreated…. It’s a fair question, but the challenge is we have no data on any type of alternate-week dosing for sacituzumab, for example, in the same way that we do with capecitabine. We don't really have that information. The sense is that it may not be as effective.

Ernst: I think that is happening in the real-world setting. I hear about it a lot. We just lack the evidence to know how to interpret the impact.

Debating UGT1A1 Testing and Primary G-CSF Prophylaxis

Leon-Ferre: If you have a UGT1A1*28 homozygous [disease], which are the ones that are poor metabolizers, how do you approach sacituzumab govitecan? Do you dose reduce? Do you do anything different?

Ernst: I’ve been following this story carefully with 2 patients since some guidance came out, particularly with testing for trial patients. [There was a] very small [number of patients], so I’m not going to let it inform completely, but it didn’t seem to pan out the way the primary guidance for dose reductions would have expected. It’s going to evolve for me. The poor metabolizers don’t seem to, in those 2 patients, have the impact I expected it would.

Leon-Ferre: Did you dose reduce?

Ernst: I did dose reduce, and then astutely, they did not have any toxicity at all, at any dose. But because of that, I was exceptionally cautious, which is very important. However, it is impactful when we’re cautious…. In these 2 patients, with liver toxicity as a potential downside effect, as well as neutropenia with sacituzumab govitecan, [but also with a high burden of disease], if we didn’t dose-adjust, it was somewhat impactful regardless of the decision. I escalated these patients cautiously to full dose. Throughout any of the exposure, there were no toxicities. Personally, I’ll have to continue to gain experience here and monitor the published data. I don’t think it is as clear as what the guidance would suggest for expected toxicity.

Batalini: That’s reflected in the data in the rates of grade 3 toxicity from both diarrhea and neutropenia. It’s not dichotomous, like having poor metabolism and having toxicity. The impact of the genotype is a little bit overrated, and we’re seeing this with not only this gene, but other metabolizer genes as well. The grade 3 neutropenia is 30%, and it goes to 40%. It’s a substantial and consistent increase, but does that 10% make us change or underdose the patient with the risk of decreasing efficacy? It’s a difficult decision. I haven’t weighed a large sum on this in my management.

Leon-Ferre: Has anyone seen someone who had this phenotype and had bad toxicity?

Ernst: I’ve certainly seen the toxicity early on in my practice, and I would like to be able to go back and characterize if those patients that I saw early toxicity from were of this phenotype. I haven’t, in this proactive setting, seen that manifest out yet.

Batalini: I had a patient who was not on sacituzumab govitecan, but she was a poor metabolizer. We found out because she had such rapid, profound, and prolonged neutropenia. It was not on sacituzumab govitecan. This was 10 years ago.

Ernst: Historically, that’s what we were associating the testing strategy to accompany: when you were seeing exceptional toxicity, consider this may be the cause. [It was a] complete flip to trying to be proactive in the way we strategize but was also potentially to the detriment of dose effects.

Leon-Ferre: I agree. I personally haven’t seen anybody with the polymorphism having bad toxicity. I’ve seen neutropenia, but it was not necessarily worse than what I would normally see.

A small study was presented at ASCO last year, which was the proactive use of both G-CSF primary prophylaxis and proactive use of loperamide [Imodium] with the first 2 cycles, with primary end points of grade 3 neutropenia and grade 2 or higher diarrhea. Long story short, lower rates were seen of both compared with the trials. There are rates of neutropenia and diarrhea, sorted by the different grades in the [phase 3 ASCENT trial (NCT02574455)], the first metastatic triple-negative setting, then the [phase 3 TROPiCS-02 trial (NCT03901339)] in the hormone receptor–positive setting, and then in the [phase 2 PRIMED study (NCT05520723)], where prophylactic G-CSF and prophylactic loperamide were used. You can see a substantial reduction of any grade, but importantly also grade 3 and 4 reductions in neutropenia and diarrhea. I don’t think this is surprising, but of course, the rates that we’re seeing in ASCENT were in the context of no prophylaxis. It was all reactive, not primary. Importantly, that is also associated with lower treatment interruptions and discontinuations.

Batalini: My bias is not to use primary prophylaxis because I don’t think adding G-CSF to these patients is trivial. We are hoping these patients are on these regimens for several months. We are having PFS that is close to a year, so we are talking about a lot of G-CSF. The flip side here in TROPiCS-02 or ASCENT is that roughly 70% of these patients during the whole duration of their time on therapy do not need sacituzumab govitecan. I personally don’t want to treat 10 patients for those 3. I’d rather identify those 3 who have neutropenia. It’s very rare with all the education we do that these patients have a severe complication from neutropenia, like grade 4. The PRIMED study was done primarily to make life easier but ultimately doesn’t meet the goal of making the patient’s life easier. Having to deal with the G-CSF, to me, is substantial. This data to me suggests not to use G-CSF and to continue being reactive. Maybe I feel different about loperamide because it’s a much more subtle intervention—easy—and you can also deescalate in later cycles. I’m not keen on the idea of primary prophylaxis G-CSF.

Leon-Ferre: To clarify, do you consider how likely there are potential risks for neutropenia? I don’t think that the recommendation is to necessarily use it on everybody, but instead to use it in those patients who have risk factors for febrile neutropenia, particularly those who are heavily pretreated, older patients. I would agree that, for example, if I have someone who has de novo metastatic disease, this is the first regimen that they’re going to be getting, or if they’re younger, with a robust marrow reserve, then it’s very reasonable not to use [G-CSF] prophylactically. Are you just talking in general, regardless of risk factors?

Batalini: I’m talking in general, but you make a very important point that where they start, and we see this in trials, if they’re starting at 1.6 [absolute neutrophil count], these are patients for whom primary prevention makes a little more sense. My practice has been just to prove that they need [G-CSF] because it’s difficult to predict after all.

Leon-Ferre: What makes me a little more worried about not using it in patients with risk factors is that we’ve seen deaths on the trials. Just like with interstitial lung disease, we’ve seen deaths with febrile neutropenia. I agree with your argument that having them be [on G-CSF] indefinitely or for however long they’re on sacituzumab govitecan is not trivial.

Case Discussion: Selecting a First-Line ADC and Managing Ocular Toxicity

Leon-Ferre: This is a 58-year-old female who is healthy overall with minimal comorbidities. She had prior stage 2 triple negative breast cancer and presented with metastatic disease 13 months after finishing her early-stage therapy, with multiple liver lesions and bilateral lung nodules. She had PD-L1 testing, and it was positive at the combined positive score of 15 and was BRCA1/2 negative. HER2 status was 0 at that point, and the [rapid plasma regain] was negative. She received [treatment from] the [phase 3 KEYNOTE-522 trial (NCT03036488) regimen], had residual disease, then received pembrolizumab alone.

In the metastatic disease setting, she presents now with a HER2 immunohistochemistry of 1+, a PD-L1 status that is now negative, and no germline mutations. Which agents would you select for first-line therapy in this patient?

If we have a show of hands for datopotamab deruxtecan (dato-DXd; Datroway), [it’s clear] that the majority favor Dato-DXd.

Has anyone had challenges with Dato-DXd toxicities, including more significant stomatitis or ocular toxicities, that make you hesitate a little bit? Or has everybody’s experience been favorable so far?

Bahadur: I don’t have a lot of experience…. but going back for this patient, I would have been fine with sacituzumab govitecan as well. Whenever I have patients with adverse effects with sacituzumab govitecan, it is better tolerated when we decrease the doses, especially with the addition of G-CSF on day 8. Either of those options would have been appropriate for this patient.

Leon-Ferre: When you see a patient in this situation, how do you discuss this with them? Do you recommend dato-DXd, forgetting about sacituzumab govitecan, or do you present both as an option? We all do shared decision-making, but as you all know, the way we present the options to the patient heavily influences what they choose. How are you approaching that discussion?

Bahadur: I do discuss both, but at the end, when you explain to them the schedule, the toxicity, and the benefit being similar, I usually say “I don’t think there's a significant difference”.

Leon-Ferre: I have to say that particularly early on, I was fairly freaked out by the ocular toxicities and we had some experience with Dato-DXd in the context of the I-SPY 2 trial in the early-stage setting... we got a little spooked by it, but we have learned how to identify those patients who are at higher risk for that. With the collaboration with ophthalmology, and also just by emphasizing the real importance of the eye drops... I’ve had patients for whom I’ve presented both options, and they really say, “ can’t wear my contact lenses?”, and I said, “Ideally not,” and then they say, “Well, I’d rather do sacituzumab govitecan”. For some patients, that's very important. There’s a space for both.