Andrew Funderburk interviews Alison Munroe, who attended this year’s American Association for Cancer Research (AACR) conference.
Andrew: Alison, what was most exciting to you at this year’s AACR?
Alison: Unsurprisingly, immunotherapy was a hot topic again at AACR, both the checkpoint inhibitors including where those will be likely be used, and cell based therapies like CAR‑T and TCR. There was a lot of interesting information about combination therapy, focused on better use of targeted therapy combinations, including better use of targeted therapy combined with chemotherapy. Sometimes we think of chemotherapy as old‑fashioned sometimes, but certainly it still does have its place in the treatment of many cancers.
Andrew: Let’s talk about the combination therapies as there is lots of news in that area. For instance, there was the simultaneous publication of the nivolumab and ipilimumab data in melanoma which had some pretty striking results as I recall.
Alison: Definitely—the efficacy levels are game changing. The response rates are also better for combinations, which is great. The side effects are also greater, however; so tolerability is a bit of a problem.
Andrew: Would you say that tolerability is the main issue with combination therapies?
Alison: Yes, especially for patients who may have existing comorbidities or who have already received multiple lines of therapy and may have lower performance status.
Andrew: What’s the sense of the oncology field’s reaction as far as the toxicity of the nivolumab-ipilimumab combination? The efficacy is stunning, but will people be able to tolerate it? Are there strategies like dose titration that can get around some of the worst toxicity issues that we see with that combo?
Alison: Possibly. I think that’s still a bit of an open question. One of the things that is nice about the combination is that ipilimumab is used for a shorter duration than the nivolumab, so the side effects profile may vary over time. The discontinuation rate due to side effects was quite high, 47% in the combination arm vs. 17% for ipilimumab monotherapy. Fortunately about 70% of those who needed to stop therapy continued to respond. Hopefully patients can still benefit from a shorter course of treatment. As far as dose modifications or how that may be modified, I think that’s still something people are looking into and requires further research.
Andrew: In the trial, what was the duration of dosing of the Ipilimumab versus the Nivolumab?
Alison: The ipilimumab is dosed every 3 weeks for 4 doses with nivolumab or placebo given at the same time points.
Andrew: Then did they continue on nivolumab?
Alison: Yes, after the 4 doses of ipilimumab, the nivolumab was continued every 2 weeks until progression or unacceptable toxicity. The majority of grade 3 and 4 adverse events were seen during the combination dosing vs. the subsequent nivolumab monotherapy. Nivolumab also looks good in single agent use. Some of decision-making for combination vs. sequential therapy may be based on whether patients have comorbidities that may make the combination more risky. The two drugs may be used sequentially if patients can’t tolerate them in combination, which I think is good news for patients.
Andrew: Yes, definitely. Let’s talk about the chemo combos and whether there are strategies using chemo or radiation. Some of these “old school” therapies could create more antigen presentation that would be beneficial for immunotherapy treatment. What did you see from AACR on that topic?
Alison: One of the areas that was particularly interesting was looking at PARP inhibitors in chemotherapy. I think the views on PARP inhibitors are getting to be much deeper as far as understanding of how they actually work and may best be used.
We call them PARP inhibitors, but their activity may not be from just inhibiting the enzyme but trapping the enzyme complex on DNA and causing DNA breaks. Using PARP inhibitors with platinum agents, specifically with carboplatin and cisplatin versus oxaliplatin seems to have better efficacy. Additionally, besides using the combination, there’s the issue of ensuring they are being used in the most beneficial patient subsets.
I think there’s some really interesting data looking at those patients beyond just BRCA1 and BRCA2 but other types of DNA repair defects. It may end up as more of a panel of genes that will be looked at rather than just one or two to decide which patients would be most likely to benefit.
Some preclinical data was presented on the order of drug dosing that suggested that a simultaneous combination, or beginning with platinum therapy was more effective than beginning with a PARP inhibitor, which resulting in lower efficacy. Research is continuing on patient subsets within different kinds of cancer and thinking more cleverly about sequential vs. combination therapies.
Andrew: It sounds like this is going to take more time and trials to really work out how exactly to use these therapies optimally.
Alison: Yes, and we are also beginning to understand drug resistance, with PARP inhibitors in some cases getting resistance from a BRCA mutant patient who may begin to re‑express a form of the BRCA protein. It’s a synthetic lethality, but then if the cancer mutates again, they can regain BRCA expression and become positive.
There was a pretty interesting plenary session on drug resistance for various mechanisms of action. Overall I think researchers are looking at resistance more proactively to enable better strategic treatment choices. There was a nice talk from Alice Shaw on the ALK inhibitors in non-small cell lung cancer (NSCLC) and cross resistance. It examined what treatments might be most effective next if a patient becomes resistant to their first line therapy. If the specific dominant mutation is identified it can help select an active second line therapy since many of the mutant forms are not cross-resistant.
Sometimes multiple drugs targeting the same mechanism are disparaged as “me-toos”, so you might question, why do we need so many ALK inhibitors? There are actually a lot of mutations which are not cross‑reactive. If an oncologist can select a treatment path using clinical evidence on mutations, patients may get through multiple lines of ALK-targeted therapy successfully and prolong response.
There was only one mutation which seemed to be cross‑reactive to multiple ALK inhibitors. Most of them are sensitive to at least a couple of them. This kind of treatment would require re‑biopsy, however, then it becomes possible to much more intelligently choose therapy.
Andrew: Right. I’m curious as to how you felt the reaction was to the CAR‑T in solid tumor data because that was eagerly anticipated. Some of the commentary I read characterized it as a bit of a swing and a miss. Granted, this is very early, and perhaps it was premature to expect real efficacy signals at this stage. Some of the liquid tumor responses in CAR‑T have been so phenomenal that people get their hopes up. We didn’t see that yet in terms of efficacy signal in solid tumor. To ask the really impossible question, where is CAR‑T going in solid tumors?
Alison: I agree the field in solid tumors for CAR-T it is just much earlier. We have to remember that. CAR‑T in liquid tumors has been making news very frequently recently, but these therapies in various iterations have been worked on for years, which sometimes gets forgotten. CD19 is the key target in liquid tumors, which been worked on for a long time as well. It’s so early in solid tumors and it is more complicated. Still, that doesn’t mean it’s never going to work, just not that it’s going to be an immediate grand-slam.
Andrew: We don’t know for sure how good of an antigen mesothelin in solid tumors is. It seems like a good hypothesis for something to go after.
Alison: For solid tumors, the choice of antigen is really important. The CAR-T construct is really important and also how accessible the tumor is. If you think about it, while liquid tumors can aggregate in the bone marrow and lymph nodes, they are, generally speaking, more accessible than solid tumors. If a solid tumor is not very vascularized, it’s much harder for a CAR‑T therapy to actually access the tumor cells. That’s where some combinations may be beneficial.
Beyond the target and actually getting good CAR-T cell access to the tumor, the dosing is also more in question. CD19 is a pretty clean target. If you kill all of the CD19-positive cells in a patient, the patient will survive. They may be a bit more prone to infections, but it’s OK, you can live without them. However, some of the other targets may be expressed at low levels on normal cells which patients can’t live without and therefore have more significant side effects. This is why the dosing may be a little bit trickier. We talk a lot about the persistence of cells, but I was speaking with KOL the other day who hypothesized, “Maybe we don’t want the CAR-T cells to persist all the time. Maybe we want them to persist for a while, kill cancer, subside and then we give another dose and then another dose over time.” I think there are a lot more questions about how all these questions will work out with the solid tumors.
Andrew: We could be looking at long tails in time of potential autoimmunity if you have perpetually persistent T cells programmed to go after targets that could be expressed at low levels on healthy tissue.
Alison: Exactly, and that would need to be managed appropriately. The goal is a long tail of efficacy, not side effects! I think that there’s just more research that needs to be done. I would characterize the data as a swing and a miss. It certainly doesn’t mean that solid tumors are out and the field should stop working.
Andrew: Maybe this was just the first pitch. You don’t usually hit a home run on the first pitch. Our expectations have gotten sky high from some of these early results.
Alison: Yes, it’s one piece of data that didn’t work. I think different targets, different methods of looking at it, even figuring out the optimal CAR‑T constructs are all being researched. I think it’s certainly possible. It’s definitely more complicated and the win didn’t happen for AACR this year.
Andrew: Anything else you thought was really noteworthy from this conference?
Alison: The FDA has been talking about how to get more patient reported outcomes (PROs) into labels. This is an ongoing conversation, continued at AACR. They are looking at better validating metrics in advance and making sure that patient reported outcome data could actually be included directly in the drug label so it is more accessible to patients. The focus is not only encouraging companies to test for patient outcomes, but making it more achievable for that data to be reportable in the label.
The FDA of course doesn’t want to lower their standards, but at the same time it is important to evolve to enable reliable clinical trial information to easily available to patients and physicians. Better ways to collect that data may make the patient survey instruments little bit less like a fishing expedition which could have be a hundred questions looking for efficacy.
If the surveys move from being paper-based to electronic, it enables them to be branching: Are you having any of these GI side effects? If the answer is no, then the patient doesn’t have to read and answer follow-up questions. Versus a patient who said, “Yes, I’m having GI side effects,” would get targeted follow up questions to characterize it more.
Andrew: It’s a much more structured way of collecting data that hopefully gets to something that’s more meaningful.
Alison: It must be more meaningful- it is a burden on patients to participate, so there must be value to ask that of patients. It’s also a burden on trial sites, but at the same time it’s really important data to have to help patients and physicians choose, especially as we have more therapies available to decide between.
That’s especially critical for therapies that can be taken for a long time. It is worth noting that while this information has the goal of helping patients, that improvements in patient experience and quality of life also can help companies differentiate their drugs. It’s good to see interest in this area from the regulatory, biopharma, and clinical researchers. The FDA is a little bit behind the EU as far as capturing patient quality of life data in a more official format.
One great point that was highlighted is the frequent delay of when the patient-reported data is analyzed and published. Obviously it’s critically important to analyze the efficacy and safety data and get the filing package to regulatory as soon as possible, so often it may not include the full analysis of PRO data. A lot of the PRO data comes out in maybe two or three papers over time after the original trial data is published. This delay may cause the information to be less impactful than if it available sooner.
Andrew: Alison, this has been a great summary and recap of AACR. We really are glad you could share some of these insights with us.
Alison: Thank you, it was great to be there at AACR 2015!
Opinions expressed here are solely those of the authors and do not reflect the views of Health Advances LLC, its management, or affiliates