IASLC World Conference on Lung Cancer
IASLC World Conference on Lung Cancer
Perspective from Robert C. Doebele, MD, PhD
September 19, 2019
4 min read

Single-agent, targeted therapies induce negligible response among smokers with NSCLC

Perspective from Robert C. Doebele, MD, PhD
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

BARCELONA — Targeted therapies for non-small cell lung cancer demonstrated potential benefit in additional molecular stratifications of the disease, according to results from the National Lung Matrix Trial presented at International Association for the Study of Lung Cancer World Conference on Lung Cancer.

However, the targeted treatments evaluated in the large precision medicine trial did not appear to benefit smokers with squamous cell carcinoma.

“We’ve known for a long time that there are certain groups of patients with lung cancer who have genomic aberrations that can be very effectively targeted; the so-called oncogene addiction … for example, the EGFR-mutant patient and RET-fusion patient. What we have in these cases is a genomically simple tumor, often in a nonsmoker, that’s got one driver,” Gary Middleton, MD, professor of medical oncology at the Institute of Immunology and Immunotherapy at the University of Birmingham in the United Kingdom, said during a press conference. “When you switch that driver off with a single agent, it works fantastically.

“What we tried to do in this study was ask a particular question: If we’ve got 20% of patients with lung cancer who have these great responses to monotherapy, how many slices of the molecular pie are there?” he added. “How many other patients are out there with specific drivers that we can successfully target? Specifically, we wanted to determine if there are any good targets in patients who smoke.”

In the umbrella phase 2 trial, Middleton and colleagues evaluated 315 patients with advanced NSCLC (median age, 65 years; 47% men). Sixty percent of the patients had stage IV disease and 24% had squamous cell lung cancer.

The researchers used a Bayesian adaptive design to screen eight targeted agents for signals of activity in 22 cohorts of patients grouped by tumor molecular characteristics.

“Using next-generation sequencing, we found out what we thought the driver might be in that particularly patient, then we matched it up to one of eight targeted therapies,” Middleton said. “What we report on today is 19 of these arms.”

Median PFS exceeding 3 months or objective response rate and/or durable clinical benefit rate greater than 30% at 24 weeks served as prespecified clinically relevant outcomes.

The target recruitment for each cohort is 30, with ongoing recruitment for 19 cohorts. The researchers reported posterior probabilities of a clinically relevant outcome for the closed cohorts and determined Bayesian predictive probability of success (PPoS) based on observed data for open cohorts.

The researchers reported outcomes from cohorts of patients treated with the cyclin-dependent kinase 4/6 inhibitor palbociclib (Ibrance, Pfizer). Results showed the following rates of median PFS: 4.5 months for patients with squamous cell carcinoma and CDKN2A loss (PPoS = .98); 3.2 months for patients with adenocarcinoma and CDKN2A loss (PPoS = .69); 2.1 months for patients with CDK4 amplification; 5.4 months for patients with KRAS mutations; 2.71 months for patients with KRAS mutations with coexisting STK11 loss; and 3.6 months for patients with CCND1 amplification.


Defined clinical benefit rates with palbociclib were 29% in squamous cell carcinoma with CDKN2A loss, 26% in adenocarcinoma with CDKN2A loss, 11% in CDK4 amplification and 19% in CCND1 amplification. The researchers predict PPoS higher than 75% in CDKN2A loss/squamous cell carcinoma and CCND1 amplification.

An analysis in which the researchers targeted PIK3CA mutations and amplifications with the AKT inhibitor capivasertib (AZD5365, AstraZeneca) showed durable clinical benefit rates of less than 15%.

In an analysis targeting lung cancer with standard oncogene addiction, crizotinib (Xalkori, Pfizer) was associated with a more than 99% PPoS in ROS1 gene fusions and MET exon 14 skipping mutations, with attenuated effects in MET amplification.

“This is what you see with oncogene addiction,” Middleton said. “We have massive levels of [durable clinical benefit] rate, very high median PFS and very high objective response rate.”

Middleton noted that further study is warranted on responses to selumetinib (AZD6244/ARRY-142886; AstraZeneca, Array BioPharma, Merck) with docetaxel among patients with NF1 mutations. Although the researchers halted recruitment to vistusertib (AZD2014, AstraZeneca) at interim for LKB1 single mutation, continuation is warranted to evaluate durable clinical benefit among patients with LKB1/KRAS double mutations.

“What does this all mean for patients with smoking-related cancers, particularly squamous cell cancers?” Middleton said. “We’ve got to think again about what we do for patients with squamous cell lung cancer. I don’t think we can copy the paradigm of single agents going for oncogene addictions. That’s not going to work.”

Middleton cited a whole-study subgroup analysis of patients according to histology and smoking status. Results showed never-smokers had an ORR of 28%, whereas smokers with 10 pack-years or less had an ORR of 17%, those with 10 to 30 pack-years had ORR of 7%, and smokers with more than 30 pack-years had an ORR of 10%.

Middleton said although there may be a good target in patients with squamous cell lung cancer, it may be necessary to shift the focus away from single-agent treatments and genomic aberrations as activators.

“We need to start thinking about the target perhaps independent of genomic aberrations activating the particular molecule and start thinking about rational combinations,” he said. “Even though we might have a good target, we have to also accept that smoking-related lung cancer is very different from breast cancer; it is very genomically complex, very unstable, and there’s an ongoing evolution of oncogene redundancy and lots of resistance.” – by Jennifer Byrne


Middleton G, et al. Abstract PL02.09. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept 7-10, 2019; Barcelona.

Disclosures: Middleton reports pharmaceutical industry partnerships with AstraZeneca, Mirati and Pfizer. Please see the abstract for all other authors’ relevant financial disclosures.