NEW YORK – Researchers at the University of Pennsylvania Perelman School of Medicine found that lung cancer patients who received liquid biopsy genomic profiling at the same time as their diagnostic bronchoscopy had shorter time to treatment than those who received the standard tissue-based testing after their first oncology visit.
The study, published in JTO Clinical and Research Reports in March, showed that the patients whose plasma samples were tested earlier using next-generation sequencing got on treatment faster. The researchers hope to explore these insights further in a larger study and may eventually implement this workflow across the institution, said Jeffrey Thompson, lead author of the study and an assistant professor of medicine at Penn Medicine.
In the pilot study of 110 advanced non-small cell lung cancer patients at Penn Medicine, researchers explored how the timing of NGS-based liquid biopsy testing impacted treatment.
Between September 2019 and April 2021, Thompson and colleagues began screening patients who were suspected of having NSCLC via imaging. At the time of their diagnostic evaluation by an interventional pulmonologist, the researchers also collected blood to be analyzed via Guardant Health's Guardant360 74-gene liquid biopsy test. The diagnostic biopsy confirmed that 55 patients had nonsquamous NSCLC as suspected and these 55 patients also went on to receive tissue-based NGS testing using a 152-gene assay, which is the standard of care at Penn Medicine.
The researchers then compared this group of patients to a retrospectively identified cohort of 55 advanced NSCLC patients at the institution who had also received standard tissue-based NGS profiling within a year of their enrollment in this study, but did not have the earlier liquid biopsy test. The medical oncologists of 16 patients in this cohort ordered plasma NGS later in the course of their treatment.
Typically, all newly diagnosed patients with advanced nonsquamous NSCLC at Penn Medicine will have their diagnostic biopsy tissue automatically sent for NGS testing, a practice called reflex tissue sequencing. This practice eliminates the need to wait for an oncologist to order tissue-based profiling when they first meet with a patient, Thompson said. However, liquid biopsy testing is often initiated after the first oncology visit for NSCLC patients.
In cohort A, an actionable driver mutation was identified in 32 patients, or 58 percent, through liquid biopsy NGS, including mutations in KRAS, EGFR, ERRB2/HER2, MET, BRAF, and RET. Tissue sequencing alone identified two additional patients with driver mutations, both harboring an EML4-ALK fusion.
In cohort A with patients who received earlier liquid biopsy testing, 85 percent of patients had liquid biopsy molecular profiling results available before their first medical oncology visit, compared to 9 percent of patients having results from tissue-based testing in cohort B, the retrospective group. While the time from the diagnostic biopsy to the first oncology visit was similar between the two groups, the tissue-testing group had a much longer turnaround time for results, 26 days compared to 8 days for the plasma NGS group. That also led to a greater proportion of cohort A patients receiving specific treatment recommendations at that first oncology visit, 74 percent versus 46 percent in the retrospective cohort B.
In this study, researchers didn't track whether patients got on a treatment matched to their tumor profiling results but they focused on how quickly they got on any treatment. Overall, cohort A patients had a shorter time to treatment because their oncologists did not have to wait as long for molecular profiling results, and had all the information they needed to quickly make therapy decisions, whether it was a standard treatment or a molecularly informed treatment.
Among all patients in the study, the median time to treatment was shorter in cohort A, 12 days, compared with cohort B, 20 days. Among the subset of patients with a specific driver mutation in both cohorts, patients in cohort A waited 10 days to get on treatment after the first oncology visit, compared to 19 days in cohort B.
The liquid biopsy workflow also helped patients who have insufficient tissue for NGS. Across both cohorts in this study, 15 percent of patients had insufficient quantity or quality of tissue DNA for NGS.
"There are a number of barriers to ensuring patients get adequate molecular profiling and having sufficient quantity or quality of DNA is certainly one of those," Thompson said. "That is one of the reasons why liquid biopsy can often be complementary to tissue [testing] in order to expedite treatment decisions."
It is standard practice for oncologists to order tissue or liquid biopsy NGS testing when patients come for their first oncology visit. However, the workflow in this pilot study, which involved ordering liquid biopsy testing at the time of their diagnostic biopsy, led to more patients being matched with targeted therapies and reduced delays in care while waiting for results, Thompson said.
"When the patient finally meets with the medical oncologist, they're told that they need to wait several weeks for molecular testing results to come back, which ultimately leads to further treatment delays and potentially inappropriate treatment assignments if treatment decisions are made prior to seeing the genotyping results," Thompson said.
Moreover, not all NSCLC patients undergo guideline-concordant biomarker testing, even though the National Comprehensive Cancer Network (NCCN) recommends testing advanced NSCLC patients with a broad, panel-based NGS assay for mutations in ALK, BRAF, EGFR, MET exon 14, NTRK, RET, and ROS1, along with testing for PD-L1 expression with an immunohistochemistry test.
A 2021 study of more than 3,400 metastatic NSCLC patients showed that 90 percent of patients were tested for at least one biomarker, but less than half were tested for all five of the most common actionable mutations in lung cancer, which were defined as mutations in ALK, BRAF, EGFR, ROS1, and PD-L1.
With the improvements in time to treatment seen in this small study, Thompson and his colleagues are exploring how to implement earlier liquid biopsy NGS for all NSCLC patients at Penn Medicine. He noted there was an "overwhelmingly positive" response to the study results and interest in implementing the workflow at the institution.
While wider implementation may take time, Thompson noted that the key to changing this workflow is clear communication. Because liquid biopsy testing is well-established at many cancer centers and is recommended by NCCN guidelines for lung cancer, getting doctors familiar with a new workflow and ensuring they know that tests have been ordered is crucial.
"Communication that these tests have been ordered and are pending, and then the clear and timely communication when those results are available to the appropriate providers, that was the key element to ensure the success of the trial," he said.
One benefit of conducting the study within Penn Medicine was that its pulmonologists, who perform the diagnostic biopsies, are well integrated within the larger thoracic oncology practice, Thompson said. Other institutions could have greater challenges adopting this type of workflow if there are more lung cancer patients referred from external clinics or if departments are more separate.
While Thompson and his team are focusing on lung cancer for now, he noted this workflow could be applicable across other tumor types. Lung cancer is a good use case for this research because of the number of targeted therapies that are available, but patients with other cancer types can also run into this "bottleneck" when they undergo tissue-based NGS, he said.
Thompson and colleagues are still exploring how to best implement this workflow clinically. In the near term, Thompson's team wants to test the impact of this workflow on patient outcomes, as well as patient and provider satisfaction, in a large, randomized study. They are also hoping to explore whether this practice might improve treatment outcomes.
"Molecular profiling of patients with newly diagnosed advanced lung cancers is critically important so that patients can be considered for personalized therapies," Thompson said. "These results show that by moving the needle forward, by performing plasma genotyping earlier in the diagnostic evaluation, you're able to have those molecular profiling results available earlier and the treatment decisions can be made earlier as well."