Frontline Immunotherapy Approval in Small Cell Lung Cancer Sets New Precedent for Progress

Anna F. Farago, MD, PhD, principal investigator for the ATLANTIS trial

Anna F. Farago, MD, PhD

Checkpoint inhibitors have propelled the field of small cell lung cancer (SCLC) forward, and further exploration into novel targets like DLL3, PARP, and activated transcription may move the needle even further, said Anna F. Farago, MD, PhD.

“The big change in 2019 is the addition of atezolizumab (Tecentriq) to carboplatin and etoposide in the first-line setting,” said Farago. “Many of the other treatments that we've been using for years are still standards, but we are starting to see more activity with immune checkpoint inhibitors, not just in the first-line setting, but in other settings as well. We'll be keeping an eye out for new approvals.”

In an interview during the 2019 OncLive^® State of the Science Summit™ on Advanced Non—Small Cell Lung Cancer, Farago, attending physician in the Center for Thoracic Cancers at Massachusetts General Hospital and assistant professor of medicine at Harvard Medical School, discussed recent advances in the field of SCLC.

OncLive: How have standards of care evolved in the field of SCLC in recent years?

Farago: SCLC is a disease that accounts for about 15% of all lung cancers; it has been a real challenge for us to treat. Generally, over the past 30 years or so, we've been using chemotherapies. In fact, the chemotherapies that we're using today are some of the same ones that we've been using for the past 30 years.

The biggest advance that we've seen in SCLC occurred over the past year, and that was a change in our first-line therapy. This change came from the IMpower133 study where investigators looked at carboplatin and etoposide, our previous standard of care, compared with carboplatin/etoposide plus atezolizumab, which is a PD-L1 inhibitor. That study showed a survival benefit for the patients who received the combination of carboplatin and etoposide with atezolizumab and has since become our new standard of care in the first-line setting. There are other randomized studies in the first-line setting that are asking similar questions with different checkpoint inhibitors. We're looking forward to seeing some of those results read out.

There are more recent studies that have been reported in the maintenance setting and in the second-line setting, such as CheckMate-451 and CheckMate-331. Unfortunately, neither one of those randomized studies was positive, but we've learned some interesting things about the activity of checkpoint inhibitors in SCLC.

[There are therapies] on the horizon for SCLC, including some single-agent immune checkpoint inhibitors in the second- and third-line settings from phase II studies, as well as some newer agents including lurbinectedin, which is an inhibitor of activated transcription, as well as PARP inhibitors and drugs that target DLL3.

How do you select patients most appropriate to receive the frontline atezolizumab regimen?

The National Comprehensive Cancer Network guidelines list the regimen as a category 1 recommended and preferred regimen. However, patients with active autoimmune diseases were excluded from the IMpower133 study; that's an appropriate exclusion. We are concerned about the risk of autoimmune disease flares or other autoimmune complications for patients who go on a checkpoint inhibitor with that medical history. That said, we do sometimes treat patients with some degree of autoimmune disorder or history, depending on how severe the event was or what the specific details were of that history.

The other population that was excluded from that study was patients with untreated brain metastases. There, we have to use our clinical judgement. It's not uncommon that patients with SCLC present with small asymptomatic brain metastases. In those scenarios, it has been common practice to start with standard first-line carboplatin and etoposide and follow the brain metastases closely. I would generally be comfortable taking a similar approach with carboplatin, etoposide, and atezolizumab for patients with small asymptomatic brain metastases at baseline. Certainly, those patients will warrant close surveillance. These are patients who will likely need whole-brain radiation at some point in their treatment.

How does frontline immunotherapy compare with frontline chemotherapy followed by maintenance immunotherapy?

That’s an interesting question. Frontline chemotherapy combined with immunotherapy was the strategy used in the IMpower133 study. Interestingly, that study showed benefit in survival and progression-free survival (PFS) with the combination, but not an improvement in response rates. When we saw that data, many in the community thought we were seeing an effect of the immunotherapy in the maintenance setting as opposed to the upfront setting with a combination of chemotherapy.

Then we saw the results of the CheckMate-451 study; this was a randomized study with 3 arms in which patients were randomized to receive maintenance ipilimumab (Yervoy) plus nivolumab (Opdivo), nivolumab alone, or placebo following 4 cycles of first-line platinum etoposide. The primary endpoint was an overall survival (OS) comparison between the ipilimumab/nivolumab arm and the placebo arm. The results were negative; there was effectively no difference in OS between those 2 groups. That came as a surprise to many of us but certainly showed, fairly convincingly, that there is not a benefit to giving ipilimumab and nivolumab in that context. Since that was the primary endpoint, the study didn't rigorously look at the statistical outcomes related to the secondary endpoints.

Interestingly, it looks like there may be a trend toward an improvement in survival with the nivolumab alone arm compared with placebo. When the investigators looked at the subset of patients who started nivolumab within 5 weeks of completing chemotherapy, there was an even greater trend toward a potential survival benefit for that group compared with those who received placebo. There are some hints there that suggest single-agent PD-1 inhibitors, or maybe in the case of atezolizumab, PD-L1 inhibitors, may confer a benefit in the maintenance setting.

However, based on the results of the CheckMate-451 study, we really don't have the data to support using checkpoint inhibitors in the maintenance setting in patients who have received upfront chemotherapy. Instead, we're favoring the combination strategy in the first-line setting and then continuing atezolizumab in the maintenance setting.

What are the big take-home messages from some of the negative studies that have read out in the space?

The CheckMate-331 study is important to discuss; that trial was a randomized study in the second-line setting. Patients who had received upfront platinum etoposide had been followed, and when their disease started to grow again, they enrolled on the study. These patients were randomized to receive either nivolumab alone or chemotherapy—topotecan or amrubicin depending on regional standards. In the United States, it's topotecan. That study was looking for an OS benefit and a PFS benefit, as they were coprimary endpoints. However, investigators did not observe a benefit in the nivolumab arm compared with the placebo arm. Interestingly, the OS looked fairly similar between the 2 groups. Perhaps there's a small subset of patients who have a prolonged survival benefit with nivolumab.

When you look at the PFS of patients who received nivolumab, the vast majority progressed very quickly. The median PFS was about 1.5 months with nivolumab compared with a little bit more than 3 months with topotecan. This tells us that the patients who are going on nivolumab and are not going to respond are going to progress very quickly. Topotecan, for all of its flaws, and there are many, is providing more disease stabilization than nivolumab.

Could you discuss the excitement with lurbinectedin?

Lurbinectedin is a really interesting drug; it's a synthetic analog of a marine-based compound called trabectedin (Yondelis), which is FDA approved for use in a subset of sarcomas. The agent is thought to function by inhibiting activated transcription. In SCLC, lurbinectedin has been studied in several phase I studies in combination with chemotherapy; that's how it was originally developed. The most notable study of the ones that have been conducted looked at the combination of lurbinectedin and doxorubicin. There were a couple of different dosing cohorts, but investigators saw high response rates ranging from 30% to [more than] 60% in 1 of the cohorts. That indicated that [lurbinectedin] certainly has some activity, albeit in a small number of patients in the second-line setting.

At the 2019 ASCO Annual Meeting, we saw some really interesting data in the second-line setting among patients who had received 1 prior line of chemotherapy; it was a single-arm phase II study in which everyone received lurbinectedin. Investigators treated a little over 100 patients and observed a response rate of about 38%, which is higher than what we see with many of the chemotherapy comparator arms with topotecan or amrubicin in many of the recent randomized studies. Investigators also observed a median PFS of 3.9 months and a median OS of 9.3 months. That OS is impressive compared with what we've seen historically in the second-line setting with chemotherapy. When you see encouraging results from a single-arm phase II study, it’s very exciting. Ultimately, we need to see randomized data. To my knowledge, there is no randomized trial planned in SCLC that is looking at single-agent lurbinectedin.

However, there is a large global randomized phase III study evaluating the combination of lurbinectedin with doxorubicin versus investigator's choice of topotecan or CAV [cyclophosphamide, doxorubicin, and vincristine]. That is a 600-patient study with a primary endpoint of OS. The trial has completed enrollment; hopefully, we'll see results within the next year. If it’s a positive study, it will likely change our practice in the second-line setting. The single-agent results that we saw at the 2019 ASCO Annual Meeting were certainly encouraging. With the single-agent study, and the previous phase I trials of lurbinectedin and chemotherapy, this drug looks like it has activity in this disease. I hope that we're going to see more development of this drug in SCLC.

How much potential does DLL3 hold as a target?

DLL3 is a negative inhibitor of the Notch signaling pathway; it is expressed on the surface of a subset of SCLC cells. Approximately 80% of patients have some expression of DLL3 on their tumor cells. This was a target that was originally identified several years ago in the development of a drug called rovalpituzumab tesirine (Rova-T). Rova-T showed really encouraging results in a phase I study; it's an antibody-drug conjugate targeted against DLL3 with a pyrrolobenzodiazepine. Unfortunately, the phase II study looking at the activity of Rova-T in the third-line setting in patients with SCLC showed a lower response rate than what we had seen previously in the phase I study.

More recently, a press release regarding the results of the TAHOE study were announced; this is a second-line randomized study comparing Rova-T with topotecan. The press release stated that this was a negative study, and, in fact, it was closed early because patients were doing worse on the Rova-T arm. We haven't seen the data yet, so it will be interesting to see some more details. There is an ongoing randomized study in the maintenance setting comparing Rova-T with placebo; we have yet to hear the results. Because the TAHOE study was negative and the results from the phase II study were underwhelming, it's hard to really know what the next steps of development with the drug would be. [The agent] also has some complicated toxicities, which can be challenging to manage clinically.

Excitingly, there are 2 drugs in development by Amgen. One is a bispecific T-cell engager, which is designed to engage DLL3 as well as CD3, which is expressed on T cells in an effort to engage the immune system against SCLC cells. The second is a CAR T-cell product targeted against DLL3; that's starting development in SCLC specifically. As such, it will be really exciting to see where these drugs go. DLL3 could still be a very viable target in SCLC.

How are PARP inhibitors being examined in SCLC?

PARP inhibitors are interesting in SCLC. They're being developed in SCLC, not because SCLC has a deficiency in BRCA or in homologous recombination, but because of preclinical data showing high expression of PARP1 in SCLC. Now, building data have shown that when given in combination with DNA-damaging agents, they can show some activity. There was a randomized phase II study looking at veliparib in combination with temozolomide compared with temozolomide alone. Although the study didn't meet its primary endpoint, which was an improvement in 4-month PFS, it did show a higher response rate with the combination compared with temozolomide alone.

At Massachusetts General Hospital, we've conducted a single-arm phase II study looking at the activity of olaparib (Lynparza) in combination with temozolomide and have also seen a fairly high response rate of over 40% in 50 patients. These are certainly encouraging results, showing some activity [with the agent]. In terms of larger randomized studies, we'll probably see those unfold over the next several years. There are also really interesting data that have come out of the Byers Laboratory at The University of Texas MD Anderson Cancer Center suggesting that PARP inhibitors may synergize with checkpoint inhibitors. We may be seeing some more interesting combinations to come.