麻豆传媒

A combination treatment for relapsed small cell lung cancer (SCLC) involving a recently approved drug failed to improve overall survival (OS) versus standard chemotherapy, though it was certainly less toxic, a phase III trial found.

In the ATLANTIS study, the median OS with lurbinectedin (Zepzelca) plus doxorubicin reached 10.6 months, statistically no better than the 9.9 months with physician's choice of chemotherapy (HR 0.967, 95% CI 0.815-1.148), reported Luis Paz-Ares, MD, PhD, of Hospital Universitario 12 de Octubre in Madrid.

"The trial didn't meet the primary endpoint, there were no differences in survival," said Paz-Ares during a press briefing at the World Conference on Lung Cancer.

While median progression-free survival (PFS) was identical at 4 months in each arm, lurbinectedin-doxorubicin still demonstrated a significant PFS improvement (HR 0.831, 95% CI 0.693-0.996). PFS rates at 6 months with lurbinectedin-doxorubicin versus controls, respectively, were 31.3% and 24.4%. At 12 months these rates were 10.8% and 4.4%.

While the average PFS was 5.9 months in the study arm and 4.6 months for the control arm, the improvement failed to translate to an OS benefit, Paz-Ares noted.

"On the more positive side, the safety profile of the experimental arm was much better, particularly in terms of myelosuppression," he said.

Grade 3 or higher hematologic adverse events (AEs) were significantly less frequent with lurbinectedin-doxorubicin versus physician's choice of chemotherapy, respectively:

• Anemia (14.5% vs 31.1%)

• Neutropenia (37% vs 69.2%)

• Febrile neutropenia (4% vs 8.3%)

• Thrombocytopenia (13.9% vs 31.1%)

Lurbinectedin is an analogue of trabectedin (Yondelis), which blocks oncogene transcription and leads to cell cycle arrest and apoptosis. Preclinical studies showed synergy between lurbinectedin and doxorubicin, Paz-Ares explained, and a phase I trial suggested every 3-week doses of 2 mg/m² for lurbinectedin (lower than the FDA-approved dose for single-agent use) and 40 mg/m² for doxorubicin -- dosing that was used for ATLANTIS.

"At the present stage, given the data we have, given the toxicity profile, I would use [lurbinectedin] single agent as compared with the combo," Paz-Ares told 麻豆传媒, noting that lurbinectedin alone has not been compared with lurbinectedin plus doxorubicin.

"ATLANTIS joins the ranks of negative phase III trials for patients with recurrent small cell lung cancer," said study discussant Charles Rudin, MD, PhD, of Memorial Sloan Kettering Cancer Center in New York City. "But in my view there are some positive outcomes here, some silver linings."

The findings offer further support for lurbinectedin's efficacy, he said. And while the combination showed some advantages over standard of care, he agreed it was best to stick with single-agent use, noting that the response rate in the trial that led to approval was numerically superior to that seen in the current study.

"The real outcome is that I think it helps to support that lurbinectedin offers a new platform on which to build better combination therapies for patients with recurrent small cell lung cancer," said Rudin.

The study randomized 613 patients (median age 63, 57% men) with relapsed SCLC to second-line treatment with lurbinectedin-doxorubicin or physician's choice of chemotherapy -- either topotecan alone, or cyclophosphamide plus doxorubicin and vincristine given at standard doses. Primary prophylaxis with granulocyte colony-stimulating factor was mandatory for the two treatment arms.

The safety profile favored lurbinectedin-doxorubicin overall, with lower rates of grade 3 or higher AEs versus controls (47.2% vs 75.4%), grade 4 AEs (16.2% vs 54.7%), and serious AEs (12.5% vs 28.7%). There was one death due to an AE in the lurbinectedin-doxorubicin arm versus 10 in the control arm.

Overall response rates were similar between the two arms, at 31.6% with lurbinectedin-doxorubicin versus 29.7% with standard chemotherapy, though responses were more sustained with lurbinectedin-doxorubicin, with a median duration of 5.7 months versus 3.8 months in the control arm (HR 0.581, 95% CI 0.416-0.812).

Baseline characteristics were well balanced between arms, with patients stratified by performance status, chemotherapy-free interval, and brain involvement, and stratification for investigator's preference for the control chemotherapy.

All patients had received prior chemotherapy and 6% had also received a PD-1/L1 checkpoint inhibitor. A little less than two-thirds were current smokers, 29%-30% were former smokers, 42%-47% had bulky disease, and 15%-16% had brain involvement. Time from diagnosis to randomization was 9 months for both arms.

Comment