Patient Selection for Eribulin in Breast Cancer

Transcript:Joyce O’Shaughnessy, MD: Where to optimally utilize eribulin varies, really, in the subtypes of breast cancer. In triple-negative breast cancer, it’s one of the very few agents we have that has been proven at the highest levels—the level-1 evidence big phase III trials—to actually be quite effective in triple-negative breast cancer. Certainly, the taxanes are effective. Platinum-based agents are effective. Gemcitabine and carboplatin have been looked at in big phase III trials. Now we have eribulin.

In Study 301 that compared second-line eribulin following an anthracycline and taxane and the second-line node status disease versus capecitabine study, they were similar in efficacy and in putting all patients together: the triple-negatives and the ER-positive patients. There were a few HER2-positive patients in there as well. But when you look at the triple-negative subset, the eribulin had a 5-month improvement in survival compared to capecitabine—9.5 versus 14.5 months’ overall survival.

Many of us don’t utilize capecitabine as a go-to drug in triple-A, except for certain circumstances where there’s a strong androgen receptor expression: bone-only, lymph node-only, that’s more luminal. There, we do utilize eribulin. Patients have bone metastases. Even if we don’t know the androgen receptor expression, we tend to utilize the capecitabine. But generally speaking, it’s not our go-to drug for the triple-negative with a lot of lung metastases or a lot of liver metastases as a single agent. Study 301 showed that eribulin is really very powerful there. It’s one of our very few level-1 evidence agents for triple-negative breast cancer.

In the ER-positive group of patients, we do have more options. There, we tend to really favor starting with capecitabine. We’re trying to minimize the toxicities and to increase the durability of response. The problem with triple-negative disease is that lifespan is so short that, unfortunately, you’ve got to get control of that disease. And unfortunately, minimizing toxicity is not usually the highest priority. It’s called getting on top of this disease and trying to prolong survival. In ER-positive breast cancer, many women will live for years, so we start off with agents such as capecitabine that have less toxicity but still very nice opportunity for durable responses. Then, we tend to go to a taxane: generally speaking, weekly paclitaxel or nab-paclitaxel because of the tolerability. It’s not myelosuppressive. You usually can, with some dose reductions if needed for neurotoxicity, keep going with that agent. You don’t have to stop it because of toxicity before the cancer has become resistant to it.

We tend to utilize those. Then, third-line, we will often utilize eribulin. However, eribulin has more myelosuppression. Sometimes, we will certainly use it if we have a heavy burden of liver metastases, but if patients have more indolent disease, a smaller burden of disease, we may hold off on the eribulin. We may utilize a single-agent gemcitabine, for example. Some patients will benefit from that. We may utilize single-agent vinorelbine—that does not cause alopecia and is a little bit less myelosuppressive than the eribulin—if we have more flexibility there, depending on the natural history and how aggressive that disease is. In HER2-positive breast cancer, there was a phase II trial I was involved in that has been published and looked at first-line eribulin with trastuzumab. It looked equally as effective as the taxanes with trastuzumab. Now though, with the addition of the pertuzumab to the CLEOPATRA regimen and the FDA approval of taxane plus trastuzumab and pertuzumab, we would not be utilizing the eribulin first-line with trastuzumab. Nonetheless, for patients that need additional proven therapeutic combinations down the line when they’ve already had, for example, pertuzumab and T-DM1 or capecitabine, the eribulin and trastuzumab combination is a combination that has proven efficacy and safety.

Mark Pegram, MD: When I think about eribulin and where it comes up in the flow of sequential use of chemotherapeutics for the treatment of advanced breast cancer, I’m really thinking about it after taxanes and after anthracyclines. Nowadays, most patients had those in their early disease setting anyway. It’s moving up in real time, because most patients are already exposed to the anthracycline and taxane classes. That’s in line with the label indication. Oftentimes in the ER-positive space, I’m using capecitabine or vinorelbine at the first foray into chemotherapeutics. Thereafter, if patients have already had prior adjuvant therapy, eribulin would come up in conversation as 1 of the preferred options. It’s usually a little bit later-line, but I have had remarkable responses and durable responses anecdotally in my practice with this drug, and I’ve used it since it became available. It is one of the mainstays in our armamentarium in 2017.

Joyce O’Shaughnessy, MD: For patients who have had quite a bit of previous neuropathy—for example, from the taxanes—but they’ve had some good recovery, if you have a choice you may not want to go back to a taxane at that time. There are some data actually showing that, by comparing ixabepilone to eribulin and looking at very careful nerve conduction studies and symptoms, eribulin was less toxic and neurotoxic than ixabepilone.

For patients who have had a lot of taxane neuropathy, the eribulin has neuropathy as well, but it’s different and sometimes patients will have less neuropathy. Preclinical studies have shown that there’s less axonal degeneration in sciatic nerve murine models, less axonal degeneration with the eribulin compared to the paclitaxel, and a little bit more upregulation of factors that will lead to neuronal growth with the eribulin versus paclitaxel. I’m not aware, however, of very good clinical data with a head-to-head comparison of paclitaxel to eribulin.

There has been a study—its accrual has been very slow; it may have been closed—comparing paclitaxel to eribulin. Hopefully, we’ll get some data. It will be interesting to look at head-to-head. That’s really what you would need to really know for sure. But from a practical standpoint in the clinic, if patients have had a lot of taxane-based neuropathy and have had some recovery, you can give them another agent that does have neurotoxicity; going with an eribulin can be less neurotoxic for them. But there is still some neuropathy, and I would expect that I might have to dose reduce over time from the standard of 1.4 mg/m2 down to the 1.2 mg/m2 or even down to 1 mg/m2 with the eribulin.

Myelosuppression is, typically, largely neutropenia. That is an issue with eribulin. We see that particularly in these more heavily pretreated patients. You sometimes do end up needing to use some hematopoietic growth factors—before the day 8 dose, for example, 1 dose or 2 of filgrastim—before the next cycle.

That is something that we commonly do, particularly in the beginning when we’re really trying to get the drug delivered. Usually, patients really need a response around that time. Over time you can certainly reduce the dose of eribulin and try to avoid the use of hemostatic growth factors, which is certainly reasonable. But early on, we’re very anxious to get the drug in and keep the patients on schedule, not be skipping doses because then you’re not going to be able to get a hold of the cancer.

We are waiting for results of a clinical trial that we conducted of every-2-weekly eribulin with pegfilgrastim. It was not required, but a lot of patients ended up getting pegfilgrastim as the day after subcutaneous injection or as the on-body patch the same day as the eribulin, every 2 weeks. It certainly was tolerable. Now we’re awaiting the safety and efficacy data.

Myelosuppression is an issue with eribulin, in some patients. Some really do very, very well and don’t have any problems at all. But it’s not uncommon to want to utilize a dose of filgrastim before the day 8 dose, and again for a dose or two before the next cycle, to try to keep people on time, particularly early on. Then, dose reduction is probably a better way to go, rather than continue to use a lot of hematopoietic growth factors.

Transcript Edited for Clarity