NEW YORK – Researchers at the University of Pennsylvania have found that non-small cell lung cancer patients who had molecular testing results prior to beginning first-line treatment lived nearly four times longer than patients who did not.
The real-world retrospective analysis, published in JCO Precision Oncology in July, included 326 newly diagnosed metastatic NSCLC patients who received care at UPenn and two of its affiliate community sites. At these sites, 80 percent of 326 NSCLC patients diagnosed between January 2019 and December 2020 had molecular testing results from analyses performed on either tissue or blood samples available before starting treatment.
Currently, the National Comprehensive Cancer Network (NCCN) recommends comprehensive molecular genotyping before first-line therapy for all newly diagnosed patients with metastatic nonsquamous NSCLC. The researchers wanted to determine whether by following this guideline oncologists could hope to improve their patients' outcomes, and they wished to explore the best testing modalities for obtaining results prior to starting treatment, said Melina Marmarelis, a study author and an assistant professor of medicine at University of Pennsylvania.
"The study was prompted because we know that molecularly informed treatment can improve survival; patients with a targetable alteration survive longer if they get a targeted therapy than if they get standard chemotherapy or immunotherapy," Marmarelis said, adding that this study confirmed "that there was a longer overall survival in the patients in the group that had the testing results available prior to first-line treatment."
Median overall survival among patients with available test results was 24.6 months compared to 6.2 months for those without results. Among 261 patients who had results, 45.4 percent had clinically actionable mutations at initial diagnosis regardless of the testing modality used, and 69.6 percent of those with actionable mutations were eligible for first-line treatments targeting those biomarkers. The most common actionable mutations found in patients were EGFR mutations followed by KRAS G12C, ALK, BRAF V600E, and ROS1 alterations.
There were several US Food and Drug Administration-approved options that NSCLC patients with actionable mutations could take in the first-line setting. In this study, most patients with EGFR exon 19, EGFR L858R, and other uncommon EGFR mutations went on to receive front-line treatment with AstraZeneca's Tagrisso (osimertinib); four patients with uncommon EGFR mutations received Boehringer Ingelheim's Gilotrif (afatinib). Eleven patients with ALK fusions, meanwhile, received Genentech's Alecensa (alectinib). Seven patients with BRAF V600E mutations received Novartis' BRAF-MEK inhibitor combination therapy of Tafinlar (dabrafenib) and Mekinist (trametinib). At the same time, patients with ROS1 fusions in this study received either Genentech's Rozlytrek (entrectinib) or Pfizer's Xalkori (crizotinib). Further, two patients enrolled in clinical trials: one with a KRAS G12C mutation enrolled in a trial of Amgen's Lumakras (sotorasib) and one patient harboring a RET fusion enrolled in a study of Eli Lilly's Retevmo (selpercatinib).
"The fastest and easiest conclusion when you see these results is … [that] if you're finding more targetable alterations, then you're prescribing more targeted therapies, and we know that those patients do better," Marmarelis said. "But we also saw benefit in patients that didn't get a targeted therapy."
Among 233 NSCLC patients without a therapeutically targetable mutation, those with pre-treatment test results had a median overall survival of 18.2 months versus 5.4 months for those without results.
Marmarelis pointed out that patients with better performance status can afford to wait for molecular profiling results, and this is probably one reason why among patients who didn't get targeted therapy, those who had pre-treatment test results still fared better than those who didn't. The latter group tended to be sicker, she said, and likely couldn't wait for results before getting on treatment. However, she noted that even after researchers adjusted for performance status, the overall survival benefit for patients who had test results persisted.
Even if test results cannot guide patients' first-line targeted therapy, the findings can provide insight into patients' prognosis or inform later lines of treatment, Marmarelis said. "There are a lot of non-targetable alterations that may inform what we do in the first line," she added.
For example, she explained, patients with mutations in STK11 or KEAP1 could be monitored more closely as these alterations are associated with poor prognosis on chemo and immunotherapy. Knowing about a KRAS G12C mutation at diagnosis can also guide the choice of initial therapy because KRAS inhibitors are currently approved for NSCLC patients who have received at least one prior systemic therapy.
Physicians used a wide range of different tests in the study, including Penn Medicine's in-house panels and commercial tests. Most patients, 71.3 percent, had both tissue and plasma testing performed.
The researchers did not report turnaround times for test results for this group of patients but noted that concurrent tissue and plasma testing was associated with a twofold higher odds of having test results before first-line therapy, likely due to the faster turnaround time of plasma testing, according to Marmarelis.
While the NCCN guidelines in NSCLC don't specify whether physicians should order tissue- or blood-based molecular testing, Marmarelis and her team suggest concurrent testing to speed up the process and ensure patients have results before they get treatment.
However, one challenge of ordering both types of NGS analyses is cost and inconsistent insurance coverage, Marmarelis said. While studies have shown the utility of both tissue and plasma testing at diagnosis, commercial insurance coverage remains a hurdle for many patients.
The Centers for Medicare & Medicaid Services has covered FDA-approved NGS panel tests since 2018 for Medicare patients with advanced cancer, but the reimbursement environment is more unpredictable for patients under commercial insurers and Medicaid. Some states, including Arizona, Illinois, Louisiana, and Rhode Island, have passed legislation covering biomarker testing, but it's not presently clear the extent to which these new laws have expanded test access.
"I hope that a paper like ours that shows an overall survival advantage to doing this type of testing [at diagnosis] would also be evidence for payors to support inclusion of concurrent testing for these patients," Marmarelis said.
The study authors are further exploring the test-associated outcomes benefit suggested by this retrospective analysis. They are enrolling metastatic NSCLC patients in a prospective study, called iNUDGE, and assessing what happens when their physicians are prompted to order molecular testing at the time of diagnosis.
The trial is using an electronic health record-based "nudge" to prompt oncologists to order both plasma and tissue-based NGS testing for patients. The researchers will measure what proportion of patients have test results prior to treatment, the turnaround time for results, the type of therapy patients received, and their overall survival.
"Another area that we continue to work in is the continuation of the care, which is also ensuring delivery of the targeted therapy," Marmarelis said. "There's a lot of education [needed] around some of the rare alterations [and] on how to sequence these targeted therapies that we're working through. [We're] trying to provide some guidance for the greater oncology community."