Precision oncology: Exciting clinical developments and complicated biology
Exciting developments in molecular oncology within the past few months include impressive clinical results of a novel drug that targets RET fusions, evidence for variability in gene mutation allele-specific therapy responses, a novel MEK1 resistance mechanism, and an emerging complexity about differences between oncogenic driver mutations regarding the biology of cancer in a given tissue of origin.
In the context of therapeutics, the idea of rechallenge in later lines of therapy is introducing new options guided by technological advances such as circulating tumor DNA (ctDNA).
Targeted therapy strikes against RET fusions
Drilon and colleagues presented results from the LIBRETTO clinical study of selpercatinib (LOXO-292, Loxo Oncology) in September at International Association for the Study of Lung Cancer World Conference on Lung Cancer. Researchers observed a remarkable objective response rate of 68% with the highly selective RET inhibitor among the first 105 primary analysis patients with RET fusion-positive non-small cell lung cancer. Researchers also observed an ORR of 91% in a subset of these patients with central nervous system metastases. All patients had received prior platinum chemotherapy. Further, an ORR of 85% occurred among 34 treatment-naive patients with RET fusion-positive NSCLC.
The results support the idea that targeted therapy is very much worth pursuing in the era of precision oncology, with RET as the latest example to demonstrate proof-of-concept in NSCLC. RET fusions are cancer drivers and, in addition to NSCLC, they occur in tumor types such as thyroid, pancreatic, colorectal and ovarian cancer, among many others.
EGFR mutation alleles affect response to checkpoint blockade
In a study published in May in Annals of Oncology, Hastings and colleagues demonstrated that although wild-type EGFR-mutated lung cancer has better response and survival outcomes after immune checkpoint therapy, different alleles of mutated EGFR respond differently. EGFR L858R (EGFRL858R) and EGFR exon 19 deletions (EGFR19) are known to have different outcomes with tyrosine kinase inhibitors.
Interestingly, in a retrospective analysis of several hundred patients from multiple cohorts, EGFRL858R responded similarly to wild-type EGFR, whereas EGFR19 had worse outcomes with anti-PD-1 or anti-PD-L1 therapy. Expression of PD-L1 was similar in the EGFRL858R and EGFR19 groups, but tumor mutational burden was lower among those with EGFR19 despite a similar smoking history. The results are intriguing and need to be validated in prospective trials.
Non-V600 BRAF-mutant tumors and anti-EGFR therapy
In another example of different driver mutations responding differentially to targeted therapy, Yaeger and colleagues found that some non-V600 BRAF-mutant colorectal cancers respond to anti-EGFR therapy. The non-V600 BRAF-mutant tumors otherwise had similar clinical characteristics. Researchers classified the mutations as class II (kinase activating and RAS independent) or class III (kinase impaired and RAS dependent).
Only one of 12 patients (8%) with class II responded to anti-EGFR therapy, whereas 14 of 28 (50%) patients with class III non-V600 BRAF mutations responded. The results are somewhat surprising, given that class III non-V600 BRAF mutations are RAS dependent, but are important in that they point to more precise use of anti-EGFR therapy for some patients with colorectal cancer who appear to have a reasonably high response rate to such therapy.
Extinguishing anti-EGFR resistance mechanisms
Cremolini and colleagues derived interesting and important observations from a prospective study of 28 patients with colorectal cancer who initially had wild-type RAS/BRAF and received anti-EGFR therapy with cetuximab (Erbitux, Eli Lilly) in combination with irinotecan-based chemotherapy in the first-line setting. After resistance emerges, such patients are treated with second-line therapy, then the question is whether there is a role for rechallenge with irinotecan plus cetuximab in the third line.
The investigators found that patients can respond to rechallenge with irinotecan plus cetuximab, but such responses appear likely to occur among patients in whom the resistance mechanisms (ie, mutant KRAS) are not detectable in ctDNA (median PFS was 4 months if mutant RAS or BRAF was not detectable in ctDNA vs. 1.9 months if detectable at the time of rechallenge).
The results are important as they offer patients the possibility of rechallenge with a measurable biomarker test. Why some patients extinguish the resistance mechanisms whereas others do not to the same extent reflects the complex nature and heterogeneity of cancer, including after removal of the selection pressure of targeted therapy.
A novel MEK1 V211D mutation
Gao and colleagues discovered that patients with BRAF K601E-mutated colorectal cancer who are treated with the allosteric MEK inhibitor binimetinib (Mektovi, Array Biopharma) plus the anti-EGFR therapy panitumumab (Vectibix, Amgen) develop a V211D mutation of MEK1 that confers therapy resistance. V211D appears to be acquired in treated tumors as it has not been observed in more than 30,000 tumors that did not receive the therapy. Preclinical data in patient-derived xenograft models suggest that an ATP competitive inhibitor of MEK1 is active against V211D-mutant MEK1.
Specific KRAS mutations
Progress in the field has involved development of KRAS G12C-specific targeted therapy that is undergoing clinical testing. Although this represents an advance, other forms of KRAS mutations remain challenging to treat. Not only does it appear there is a need to target specific KRAS mutations using different therapeutic approaches, but also it is becoming clear that different KRAS mutations have different biology, a different proteome and different interactions in tumors that arise in different tissues.
A report in Cancer Discovery by Poulin and colleagues showed that, in preclinical models of colon cancer, different KRAS mutations drove disease that had different biology in terms of aggressiveness. The investigators engineered mice with KRASG12D or KRASA146T alleles that appeared to exhibit distinct tissue-specific effects that they attributed to allele-specific and context-dependent variations in signaling by the various mutants. This is further evidence for complexity that must be considered in therapeutic design.
Cremolini C, et al. JAMA Oncol. 2019;doi:10.1001/jamaoncol.2018.5080.
Drilon A, et al. Abstract PL02.08. Presented at: International Association for the Study of Lung Cancer World Conference on Lung Cancer; Sept. 7-10, 2019; Barcelona.
Gao Y, et al. Cancer Discov. 2019;doi:10.1158/2159-8290.CD-19-0356.
Hastings K, et al. Ann Oncol. 2019;doi:10.1093/annonc/mdz141.
Poulin EJ, et al. Cancer Discov. 2019;doi:10.1158/2159-8290.CD-18-1220.
Yaeger R, et al. Clin Cancer Res. 2019;doi:10.1158/1078-0432.CCR-19-2004.
For more information:
Wafik S. El-Deiry, MD, PhD, FACP, is associate dean for oncologic sciences at Warren Alpert Medical School and director of the Joint Program in Cancer Biology at Brown University, as well as HemOnc Today’s Associate Editor for Molecular Oncology. To contribute to this column or suggest topics, email him at email@example.com.
Disclosure: El-Deiry reports no relevant financial disclosures.