ASCO 2020: Key Lung Cancer Studies

CME

Key Studies in Lung Cancer: Independent Conference Coverage of the ASCO 2020 Scientific Meeting

Physicians: Maximum of 1.25 AMA PRA Category 1 Credits

Released: September 25, 2020

Expiration: September 24, 2021

Stephen Liu
Stephen Liu, MD
Zofia Piotrowska
Zofia Piotrowska, MD

Activity

Progress
1
Course Completed
Immunotherapy in SCLC

Stephen V. Liu, MD
We will now switch gears and discuss important updates in SCLC that were presented at ASCO 2020.

Coming into ASCO 2020, we currently have 2 standard-of-care immunotherapy options for the treatment of newly diagnosed extensive-stage SCLC: atezolizumab and durvalumab, each in combination with chemotherapy.41,42 Results from 2 separate randomized phase III trials (IMpower133 and CASPIAN, respectively) demonstrated that the addition of these PD-L1 antibodies to chemotherapy improved survival in NSCLC, leading to their approval by the FDA in this setting.41-44

At ASCO 2020, we saw reports from 2 additional trials evaluating first-line chemoimmunotherapy for extensive-stage SCLC, but now focused on the addition of the PD-1 antibodies pembrolizumab or nivolumab to chemotherapy.

KEYNOTE-604 Final Analysis: Addition of Pembrolizumab to Etoposide and Platinum in Extensive-Stage SCLC

Stephen V. Liu, MD
The first of these trials was KEYNOTE-604, which was a randomized, double-blind phase III trial evaluating the addition of pembrolizumab to first-line platinum and etoposide chemotherapy in patients with extensive-stage SCLC (N = 453).45,46

All patients received platinum-based chemotherapy with carboplatin (area under the curve 5) or cisplatin 75 mg/m2 on Day 1 plus etoposide 100 mg/m2 on Days 1-3 every 3 weeks for 4 cycles. Patients were randomized to receive either concurrent pembrolizumab 200 mg or placebo on Day 1. After 4 cycles, pembrolizumab and placebo were continued as maintenance therapy every 3 weeks for up to 31 cycles. Patients who experienced a CR or PR after 4 cycles of maintenance therapy were eligible to receive up to 25 Gy of prophylactic cranial irradiation at the investigator’s discretion.

The dual primary endpoints were PFS and OS. Secondary endpoints included ORR, duration of response, and safety.

KEYNOTE-604: PFS in ITT Population

Stephen V. Liu, MD
Addition of pembrolizumab to chemotherapy significantly improved PFS, with a median PFS of 4.5 months vs 4.3 months with placebo (HR: 0.75; 95% CI: 0.61-0.91; P = .0023). In the final analysis, the median PFS was 4.8 months vs 4.3 months, respectively (HR: 0.73; 95% CI: 0.60-0.88).

KEYNOTE-604: OS in ITT and As-Treated Populations at Final Analysis

Stephen V. Liu, MD
However, this study did not meet its OS endpoint. Although there was a numeric trend for a modest improvement in OS with the addition of pembrolizumab to chemotherapy (median OS ITT: 10.8 vs 9.7 months; HR: 0.80; 95% CI: 0.64-0.98; P = .0164), it did not cross the predefined statistical threshold for positivity (1-sided P = .0128). Therefore, this study did not show a statistically significant OS benefit for this regimen.

KEYNOTE-604: Safety

Stephen V. Liu, MD
The overall safety profile for pembrolizumab plus chemotherapy in KEYNOTE-604 was favorable and comparable with previous reports. The overall incidence of all-cause AEs was similar between the pembrolizumab and placebo arms (any grade: 100% vs 99.6%, respectively; grade 3/4: 76.7% vs 74.9%, respectively). However, more patients on pembrolizumab discontinued any therapy due to AEs: 14.8% of patients in the study group and 6.3% of those in the placebo group. As expected, there was a higher incidence of immune-related AEs with pembrolizumab compared with placebo.

KEYNOTE-604: Clinical Implications

Stephen V. Liu, MD
Overall, it was somewhat disappointing to not see a significant survival benefit with the addition of pembrolizumab to chemotherapy in patients with advanced SCLC. Although the outcomes with this intervention were in line with those seen with atezolizumab and durvalumab in the IMpower133 and CASPIAN trials, respectively, this trial simply did not meet its survival endpoint and, in my opinion, the results do not change practice.

Dr. Piotrowska, were you surprised by these results?

Zofia Piotrowska, MD, MHS
My take is that it was a difference in the statistical design that led to KEYNOTE-604 not meeting the threshold of significance for an OS benefit. Numerically, the modest OS benefit seen in KEYNOTE-604 was quite similar to what we saw in both IMpower133 and CASPIAN,43,44 and even though I agree this study won’t change my practice, the data are consistent with the benefit of adding immunotherapy to chemotherapy in that first-line treatment of extensive-stage SCLC that was previously demonstrated.

Stephen V. Liu, MD
I agree. These data do reinforce the strategy of combining an immune checkpoint inhibitor with chemotherapy in this setting, and although unlikely to lead to an approval, I would be comfortable building on these results in future studies.

Which of the following findings was reported with first-line pembrolizumab plus etoposide and platinum chemotherapy vs placebo plus etoposide and platinum chemotherapy in patients with extensive-stage SCLC in the phase III KEYNOTE-604 trial?
Phase III CASPIAN: First-line Durvalumab With and Without Tremelimumab Plus EP vs EP Alone in Extensive-Stage SCLC

Stephen V. Liu, MD
CASPIAN was an open‑label, active-controlled, randomized, phase III trial evaluating the addition of the PD-L1 antibody durvalumab with or without the CTLA-4 antibody tremelimumab to platinum-based chemotherapy in patients with treatment‑naive extensive‑stage SCLC (N = 805).47

Chemotherapy was given as etoposide 80-100 mg/m2 on Days 1-3 plus either carboplatin area under the curve of 5-6 or cisplatin 75-80 mg/m2 on Day 1 every 3 weeks for 4-6 cycles. Durvalumab was given at 1500 mg every 3 weeks, and tremelimumab was dosed at 75 mg every 3 weeks. The chemotherapy-alone arm could receive optional prophylactic cranial irradiation as maintenance and up to 6 cycles of chemotherapy. Both experimental arms received only 4 cycles of chemotherapy followed by maintenance durvalumab monotherapy every 4 weeks until disease progression. The primary endpoint is OS, and secondary endpoints include PFS, ORR, safety/tolerability, and patient-reported outcomes.

As mentioned above, previously reported data from CASPIAN demonstrating a significant OS benefit with combination therapy (HR: 0.73; 95% CI: 0.59-0.91; P = 0.0047) led to the approval of durvalumab plus chemotherapy by the FDA in early 2020 as frontline treatment for extensive-stage SCLC.42,44

At ASCO 2020, Paz-Ares and colleagues presented results from the third arm, where both durvalumab and tremelimumab were added to chemotherapy along with updated OS results for durvalumab plus chemotherapy with an additional 11 months of follow-up.47

CASPIAN: OS for Durvalumab Plus Tremelimumab Plus EP vs EP

Stephen V. Liu, MD
Unfortunately, the combination of durvalumab and tremelimumab plus chemotherapy was no better than chemotherapy alone, with no significant difference in the median OS (10.4 vs 10.5 months, respectively; HR: 0.82; 95% CI: 0.68-1.00; P = .0451; statistical significance required P < .0418).

CASPIAN: OS for Durvalumab Plus EP vs EP

Stephen V. Liu, MD
Updated data for the combination of durvalumab plus chemotherapy continued to show an OS benefit with longer follow-up, which was reassuring. The updated median OS for the durvalumab group was 12.9 months vs 10.5 months for chemotherapy alone (HR: 0.75; P = .0032). OS rates at 12 months were 52.8% vs 39.3%, respectively; at 18 months, 32.0% vs 24.8%; and at 24 months, 22.2% vs 14.4%.

CASPIAN: PFS for Durvalumab Plus EP vs EP

Stephen V. Liu, MD
There was also a trend toward improvement in PFS with durvalumab plus chemotherapy, but this was not formally tested due to the hierarchical design of this study. Landmark PFS rates at 12 months were 17.9% vs 5.3%, respectively; at 18 months, 13.9% vs 3.4%; and at 24 months, 11.0% vs 2.9%.

CASPIAN: OS in All Arms

Stephen V. Liu, MD
In summary, the addition of durvalumab to chemotherapy improved OS, whereas the addition of both durvalumab and tremelimumab tracked more closely to chemotherapy alone, at least initially. Although the survival curves for both arms with durvalumab eventually aligned, it is clear that tremelimumab did not offer a meaningful improvement in OS.

CASPIAN: Safety

Stephen V. Liu, MD
What tremelimumab did add, unfortunately, was toxicity, with 70.3% of patients experiencing grade 3/4 AEs vs 62.3% with durvalumab and chemotherapy. Discontinuation due to AEs was seen in 21.4% of patients receiving both antibodies, which is double that with durvalumab alone at 10.2%. The addition of tremelimumab also doubled treatment‑related deaths: 10.2% vs 4.9% with durvalumab plus chemotherapy vs 5.6% with chemotherapy alone. The rate of immune-mediated AEs was higher with dual-antibody treatment as well, at 36.1% vs 20.0% with durvalumab plus chemotherapy and 2.6% with chemotherapy alone.

CASPIAN: Clinical Implications

Stephen V. Liu, MD
Although there was a preclinical rationale for the hypothesis that the addition of a CTLA-4 antibody to upfront chemoimmunotherapy with a PD-L1 inhibitor would improve outcomes, we simply did not see that translate into clinical benefit. It did increase toxicity and, in my mind, does not change our current practice. It also raises questions as to whether there is a role for targeting CTLA-4 in the treatment of SCLC.

Zofia Piotrowska, MD, MHS
This was a disappointing result for the combination of durvalumab plus tremelimumab, and particularly considering the increased toxicity, has solidified that this combination does not have a role in the current treatment landscape for SCLC. That said, I agree that it was reassuring to see durvalumab plus chemotherapy continuing to show benefit in longer follow-up, and on par with what we would expect based on similar studies—which really is the practice-changing component of the CASPIAN study.

Lurbinectedin in SCLC: Background

Stephen V. Liu, MD
Finally, we will look at 2 additional studies on options for patients with relapsed SCLC after platinum-based chemotherapy.

The first is lurbinectedin, a novel transcription inhibitor that was granted accelerated FDA approval in June 2020 for metastatic SCLC with progression during or after platinum-based chemotherapy.48 Approval was based on a phase II, multicenter, open-label, basket trial of lurbinectedin monotherapy 3.2 mg/m2 in 105 patients with relapsed NSCLC that demonstrated an ORR of 35.2% at a median follow-up of 17.1 months. The ORR was 22.2% in patients with platinum‑resistant disease and 45.0% in those whose disease was platinum sensitive.49 

Phase Ib/II Trial of Lurbinectedin Plus Irinotecan in Patients With Advanced Solid Tumors

Stephen V. Liu, MD
At ASCO 2020, Aix and colleagues50 presented results from an ongoing, multicenter, open-label phase Ib/II trial of lurbinectedin plus irinotecan in multiple tumor types (planned N = 150). Irinotecan is an attractive partner because it is active in SCLC and other tumor types.

Following completion of phase Ib dose-escalation and determination of the recommended phase II dose, patients were treated with lurbinectedin 2.0 mg/m2 on Day 1 plus irinotecan 75 mg/m2 on Days 1 and 8 every 3 weeks, in phase II. Secondary endpoints included safety, pharmacokinetics, responses, PFS, and OS.

Lurbinectedin Plus Irinotecan in Advanced Solid Tumors: Safety

Stephen V. Liu, MD
At the time of this report, 59 patients enrolled in phase II were evaluable for safety. Unfortunately, this combination was somewhat toxic, with fairly high rates of grade 3/4 myelosuppression (49% neutropenia, 13% anemia, and 10% thrombocytopenia). Nearly 30% of patients required dose reductions, 76% of which were due to omission of Day 8 irinotecan due to AEs.

These data are due to the largely overlapping toxicity of both lurbinectedin and irinotecan, particularly in more heavily pretreated tumor types, such as SCLC and endometrial cancer.

Lurbinectedin Plus Irinotecan in Advanced Solid Tumors: Efficacy

Stephen V. Liu, MD
With regard to efficacy, the response rate in SCLC (n = 13) was high at 69.2% (all PRs), but there was little signal in the other tumor types. The median PFS of 4.3 months in patients with SCLC was somewhat modest.

Lurbinectedin Plus Irinotecan in Advanced Solid Tumors: Clinical Implications

Stephen V. Liu, MD
I’m not sure that adding irinotecan to lurbinectedin will be a combination worth developing further, but I am interested to see more mature results from this relatively small study.

What do you think, Dr. Piotrowska?

Zofia Piotrowska, MD, MHS
It is exciting to have lurbinectedin as a treatment option for our patients with relapsed SCLC after chemoimmunotherapy, and I was excited about the recent FDA approval. I agree, however, that combining with irinotecan might not be the best way to use lurbinectedin. Results of lurbinectedin monotherapy seem to be quite good so there may be no need for chemotherapy as a partner.

Phase II RESILIENT Part 1: Liposomal Irinotecan After Platinum Chemotherapy in SCLC

Stephen V. Liu, MD
RESILIENT is an ongoing, open‑label phase II/III study of liposomal irinotecan vs topotecan for SCLC with progression on or during platinum-based chemotherapy or chemoradiotherapy.51 At ASCO 2020, Spigel and colleagues presented results from a post hoc subgroup analysis of the single-arm, part 1, phase II portion of RESILIENT.52 This report encompassed 25 patients who received the recommended dose of liposomal irinotecan 70 mg/m2 every 2 weeks, and stratified patients by platinum sensitivity or resistance (defined as disease progression after or within 90 days of platinum-based chemotherapy, respectively). Asymptomatic CNS metastases at the time of enrollment was permissible.

RESILIENT Part 1 Subgroup Analysis: Response

Stephen V. Liu, MD
What we saw in this relatively small study was an intriguing ORR of 44% (all PRs), including a 30% rate in the platinum‑resistant cohort, which is quite high for this group, and a 53% rate for the platinum‑sensitive cohort.

These are encouraging responses, but this is a fairly small trial. It will be interesting to see the results in the larger phase III trial that is Part 2 of this study.

RESILIENT Part 1 Subgroup Analysis: Safety

Stephen V. Liu, MD
There was some toxicity with this combination—100% and 40% of patients had an any-grade or grade ≥ 3 treatment-emergent AE, respectively. Although liposomal irinotecan is a novel delivery strategy, there were notable rates of grade 3 diarrhea (20%) and neutropenia (16%). Two patients, both in the platinum-sensitive cohort, died from abdominal sepsis related to irinotecan.

RESILIENT Part 1 Subgroup Analysis: Clinical Implications

Stephen V. Liu, MD
The high response rate seen with this formulation of irinotecan is encouraging, but it is important to see if those numbers are confirmed in larger studies. Overall, it’s certainly good to have more active drugs in development for SCLC. However, what we really need is new drugs with durable and meaningful responses and biomarkers to guide their implementation into practice.

Dr. Piotrowska, what are your thoughts about this study?

Zofia Piotrowska, MD, MHS
A new formulation of irinotecan that decreases the toxicity burden would be clinically meaningful, but I also am more interested in novel agents that, as you mentioned, are biomarker driven. Furthermore, I would prefer to reserve irinotecan for later lines of therapy and improve the first-line and second‑line options before we get there.