Should We Be Doing Molecular Testing at the Time of Acquired Resistance?

H. Jack West, MD


April 10, 2019

Repeat biopsies, done throughout the evolution of a cancer and its development of acquired resistance, represent a moving target in managing lung cancer as well as other primary cancers.

Increasingly we expect to see molecular correlates as a cancer transitions from sensitivity to resistance to a given systemic therapy. This growing understanding can lead directly to new treatment directions. In patients with advanced non–small cell lung cancer (NSCLC) that harbors an activating EGFR mutation and has progressed on a first- or second-generation EGFR tyrosine kinase inhibitor (TKI), finding the EGFR T790M mutation on a repeat biopsy using either progressing cancer tissue or a blood-based test suggests a strong probability that the patient will respond to the third-generation EGFR TKI osimertinib.[1] In this setting, a repeat biopsy is a gateway to changing management. But many leading cancer researchers favor performing a repeat biopsy with the hope of finding a new, actionable mutation or providing insight into the underlying biology of cancer progression under the selective pressure of an active systemic therapy.

Of note, as alectinib or other second-generation ALK inhibitors become the clear first-line standard for patients with ALK-positive NSCLC, the molecular patterns underpinning acquired resistance also change, with approximately 50% of patients exhibiting an ALK mutation at the time of progression.[2] Lorlatinib has what may be unique activity in patients with extensively pretreated ALK-positive NSCLC, with a response rate of 39% and progression-free survival of 6.9 months in patients who received two or more prior ALK inhibitors—treatment that would include at least one second-generation agent.[3] This activity has led the US Food and Drug Administration to approve lorlatinib for patients with advanced ALK-positive NSCLC "whose disease has progressed on crizotinib and at least one other ALK inhibitor for metastatic disease or whose disease has progressed on alectinib or ceritinib as the first ALK inhibitor therapy for metastatic disease."[4]

In this setting, Alice Shaw, MD, PhD, and colleagues—a group of international all-stars on ALK-positive NSCLC—just published a potentially practice-changing study on the value of repeat tissue biopsy or plasma genotyping with the Guardant 360 assay to predict the probability of a patient responding to the newly approved, later-generation ALK inhibitor lorlatinib.[5] The analysis reviewed tissue and plasma-based testing of circulating tumor DNA (ctDNA) at the time of entry onto a prior clinical trial of lorlatinib among 198 patients who had received prior ALK inhibitor therapy and had been enrolled on one of the trial's expansion cohorts. Among the subset of 76 (78%) samples that were sufficient for genotyping among 98 de novo (postprogression) tissue biopsies, the rate of detection of ALK mutations was 47%. Plasma genotyping was done on 189 postprogression cases, with a sensitivity of 61% and specificity of 82% compared with tissue as a benchmark, with concordance of plasma and tissue genotyping of 73%. There were different patterns of specific mutations observed depending on prior drug exposure, most notably that the G1202R mutation was detected in the majority of patients who received one or more second-generation ALK inhibitors (53% and 55% of plasma and tumor tissue cases, respectively).

This is not a test that should change management, regardless of the result.

Arguably, the most relevant finding was that while the response rate to lorlatinib did not differ significantly on the basis of ALK mutation testing results among patients who developed resistance to crizotinib alone, those who developed acquired resistance to one or more second-generation ALK inhibitors and who had one or more ALK mutations were significantly more likely to demonstrate a response to lorlatinib than those who did not. By plasma genotyping, where 34 (26%) of 128 patients had an ALK mutation detected, the objective response rate (ORR) was 62% versus 32% for those with or without an ALK mutation detected, respectively; results were very similar for those defined as ALK mutation positive versus negative by tissue testing (69% vs 27%, respectively).

Interesting to note is that results for PFS and duration of response (DOR) were similar regardless of the presence or absence of an ALK mutation by plasma testing, but both PFS and DOR were significantly longer in ALK mutation-positive patients as defined by tissue testing (11.0 months vs 5.4 months and 24.4 months vs 4.3 months for PFS and DOR, respectively).

Where are we left in the wake of these results? Should they compel us to perform molecular testing at the time of acquired resistance?

Though the results show that the presence of an ALK mutation at repeat biopsy after progression on a second-generation ALK inhibitor—whether from tissue biopsy or plasma ctDNA—predicts about twice the probability of an objective response in patients without an ALK mutation, this finding doesn't necessarily change management. Unlike the situation with osimertinib for patients with EGFR mutation-positive NSCLC and T790M-negative acquired resistance after prior first- or second-generation EGFR TKI therapy, in which osimertinib has minimal and only very transient activity in those without a T790M mutation, the ORR to lorlatinib exceeds that of other alternatives in this setting, even among those patients without an ALK mutation detected. This is not a test that should change management, regardless of the result.

There are other reasons to do a test, including potentially refining expectations, and especially to improve our understanding of the biology of this treatment setting. The discordant results seen between plasma and tissue testing should remind us that, despite the promise of plasma genotyping and many early successes, it is too early to presume that an easily obtained blood-based test is completely interchangeable with a repeat tissue biopsy. We need further study of whether or when plasma genotyping is an acceptable and sufficiently reliable alternative.

In the end, when there is no clear clinical need to pursue a test, it rightly falls to the judgment of an oncologist and patient about whether to pursue a test primarily for greater personal insight and as a contribution to the field. Many patients, particularly those in a narrow molecularly defined subgroup, feel a kinship with other patients and want to facilitate progress in understanding their cancer. Whether that merits accepting the financial cost of a test that may not be covered as a necessary intervention is an individual decision. In the meantime, today this approach is a means to accelerate research, even if it cannot be considered an expectation—and certainly not a mandate—for these patients.

I welcome your thoughts. Do you encourage a patient to pursue tissue and/or plasma testing to obtain a result that may offer insight but that isn't critical to modify clinical decisions?

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.