NEW YORK – New data this month has bolstered the evidence that HER2-targeting drugs can be effective against a variety of different tumors in patients with HER2 amplification.
Tracking advanced cancer patients who received next-generation sequencing as part of normal clinical practice, investigators from MD Anderson Cancer Center reported that significant numbers of individuals had HER2 amplification across tumor types for which anti-HER2 drugs are not currently approved. And in the subset of patients who were referred to various clinical trials investigating HER2 targeted therapy, drug responses suggested that they derived more benefit from these treatments than they had with prior non-targeted treatment.
Although this comparison wasn't randomized, the study's authors wrote that their intra-patient comparison of current HER2 drug response versus prior lines of treatment, coupled with time-dependent analysis of overall survival in those who received targeted treatment versus those who did not, suggests patients may be deriving real benefit from the HER2-targeted approach.
The US Food and Drug Administration has approved HER2 targeted treatments for HER2 overexpressing breast cancer and metastatic gastric cancer.
But basket trials have made it easier for drugmakers and researchers to identify efficacy signals for targeted drugs and treatment combinations across different molecularly defined tumor types. HER2 targeted therapies have sparked interest because they have shown efficacy in different tumor types. For example, the MD Anderson authors wrote that following positive results in colorectal cancer patients in the ongoing Phase II basket trial MyPathway, the National Comprehensive Cancer Network updated its guidelines to endorse pertuzumab plus trastuzumab (Genentech's Perjeta plus Herceptin) and trastuzumab plus lapatinib (Novartis' Tykerb) in HER2-positive colorectal cancer.
While some oncologists may be prescribing HER2-targeted therapy in tumor types other than breast and gastric cancer based on emerging evidence, regulatory approval of a pan-cancer indication will require more research. Some experts in the field even doubt that a pan-cancer approval for a HER2-targeted drug is feasible given the relative rarity of HER2-positive patients across an unspecified population.
Nonetheless, researchers are continuing to investigate the possibility. In their study this month, investigators sought to provide an updated readout of both HER2 amplification prevalence and drug response. Overall, the team studied 5,002 patients with advanced cancer (excluding breast cancer) who underwent NGS as part of MD Anderson's precision oncology practice between January 2011 and June 2017. Within that group, they tracked the overall prevalence of HER2 amplification and the percentages for each cancer type represented in the cohort.
They also measured the correlation between sequencing-based HER2 testing — a measure of copy number changes in the gene ERBB2 — against immunohistochemistry and FISH methods, which have been the mainstay for assessing the biomarker in breast cancer, as well as liquid biopsy NGS results for a subset of patients tested by Guardant Health.
According to the authors, 122 patients (2.4 percent of the overall cohort) had HER2 amplification, including more than 5 percent of patients with endometrial and bladder cancers, about 5 percent of biliary and gallbladder cancers, 4.7 percent of salivary cancers, and 3.6 percent of colorectal cancers.
"This is a meaningful percentage of patients," said Ecaterina Dumbrava, the study's first author and an assistant professor in MD Anderson's department of investigational cancer therapeutics.
Examining HER2 drug response, the researchers then excluded patients from the cohort with gastric and gastro-esophageal cancers, which represent the highest non-breast prevalence.
That left 40 individuals who were referred to a variety of anti-HER2 drug trials and received at least one line of targeted treatment. Most (93 percent) received trastuzumab in combination with chemotherapy or other targeted therapies. Across different lines of treatment, 27 patients received trastuzumab with other targeted therapies such as pertuzumab, 14 patients received trastuzumab and chemotherapy, three patients received small-molecule inhibitors targeting HER2, three patients received antibody-drug conjugates or bispecific antibodies against HER2, and two patients received trastuzumab alone.
Across these individuals, median overall survival was 18.6 months on HER2-targeted treatment, compared with 10.9 months for non-targeted treatment, which the authors described as a clinically, if not statistically, significant result. "Without question for us that is meaningful, even if it has marginal significance statistically," Dumbrava said.
For progression-free survival, the group compared PFS seen on-treatment with the HER2 regimen a patient had against the PFS they experienced on whatever their last treatment had been. This measurement allows each patient to be their own control, Dumbrava explained. It is generally accepted in the field that when the resulting ratio is 1.3 or greater, the patient has derived a meaningful benefit from the second treatment.
This PFS ratio was at least 1.3 for more than half the patients in the present study. Median HER2 PFS was 24 weeks, compared to just 13 weeks on average for prior treatment regimens.
"This is a significant increase in PFS," Dumbrava added. "It doubled almost, which is really impressive. And considering that HER2 targeted therapy is usually well-tolerated, I think this is a really important finding," regarding the potential benefit for patients beyond the breast and gastric populations.
According to the authors, 12 of the 40 non-gastric cancer patients saw their tumors shrink according to RECIST criteria. Seven of these patients received trastuzumab and pertuzumab, four got trastuzumab with chemotherapy, and one patient was treated with an HER2 antibody-drug conjugate. Another nine patients had RECIST-defined "stable disease" for at least 24 weeks.
Of the trial subjects who tested positive for HER2 amplification, 68 didn't get HER2-targeted therapy. According to Dumbrava and colleagues, the leading cause for this was non-eligibility for a clinical trial. Having equivocal test results also contributed, as did insurance denials, and clinic issues facing patients with advanced and progressing disease.
Also important in the study was the opportunity to compare HER2 status as determined by different platforms. As the use of NGS continues to increase among advanced cancer patients, HER2 amplification will be more and more frequently noted in the context of sequencing, as opposed to the FISH and IHC methods that more commonly serve as companion diagnostics for approved drugs.
In the MD Anderson cohort, 42 of the 122 patients identified as having HER2 amplifications via NGS also underwent IHC testing. Of these, 31 showed HER2 protein overexpression, while four had equivocal expression, two had low expression, and five showed no HER2 protein expression at all.
Sixteen patients had testing by FISH in addition to NGS. In fourteen of these (about 88 percent) FISH results were concordant with NGS. Authors wrote that a large number of the discordance cases could be explained by the presence of equivocal, or borderline NGS results. Dumbrava also noted that NGS classification of HER2 can vary from one NGS platform to another, something that clinicians may have to grapple with in the future.
Twenty-four patients in the study, for example, tested positive for HER2 using the Guardant360 liquid biopsy test, which calculates HER2 cutoffs differently than tissue NGS. As long as these results were "strong," or "very strong" positives, they correlated with IHC and FISH.
Dumbrava highlighted the recent changes in NCCN guidelines for colorectal cancer, based on MyPathway results, as an example of how the field is expanding use of HER2-targeted therapies, even if these drugs end up not being approved for tissue-agnostic or pan-cancer use.
"Personalized cancer treatment is becoming more tissue-agnostic and we are choosing treatment more based on genomics," Dumbrava said. However, "the path for approval, is just cleaner … easier, if it's in a specific tumor type."
That said, she remains hopeful that HER2-targeted drugs will be approved regardless of tumor type, especially with results emerging like the ones from the present trial. Although HER2 amplification tends to occur relatively rarely in tumors beyond breast and gastric cancers, the benefit seems to be promising enough that these responses should be further evaluated, Dumbrava said.