NEW YORK – Researchers are following up on a finding that suggests colorectal cancer patients with high HER2 expression may fare better on anti-EGFR treatment, counter to earlier research characterizing its expression as a potential resistance biomarker.
In a study published in the Journal of Clinical Oncology earlier this month, researchers found that colorectal cancer patients' HER2 expression levels may have influenced whether they benefited on first-line treatment with chemotherapy plus either Eli Lilly's EGFR inhibitor Erbitux (cetuximab) or Genentech's VEGF inhibitor Avastin (bevacizumab).
The researchers analyzed gene expression and treatment outcome data from more than 900 colorectal cancer patients enrolled in the Phase III Cancer and Leukemia Group B (CALGB)/SWOG 80405 trial. In that trial, patients received either Erbitux and chemo or Avastin and chemo. According to the trial's previous data readouts, which did not include analysis of outcomes by patients' HER2 expression status, there was no significant difference in overall survival when Erbitux or Avastin was added to first-line chemotherapy.
Erbitux was approved in the US as a first-line treatment for KRAS wild-type colorectal cancer in 2012, while Avastin was approved for an all-comer colorectal cancer population in 2004.
Knowledge about which patients benefit most from anti-EGFR therapies and understanding of the role of HER2 expression in colorectal cancer have evolved since the CALGB/SWOG 80405 trial began in 2005, said Heinz-Josef Lenz, corresponding author of the newest study and associate director for clinical research and co-leader of the gastrointestinal cancers program at the University of Southern California's Norris Comprehensive Cancer Center.
"We were interested in HER2 because it is a very novel target for drug development for colorectal cancer, with the recent approval of tucatinib for colorectal cancer patients who have an amplified, overexpressing HER2 gene," Lenz explained.
Seagen's HER2 inhibitor Tukysa (tucatinib) plus Genentech's HER2 antibody Herceptin (trastuzumab) was approved in the US last year for metastatic colorectal cancer patients whose tumors are RAS wild-type and HER2-positive, based on a study showing a 38 percent response rate and a median duration of response of 12.4 months. Other recent studies have also shown that HER2 overexpression is an efficacy-associated biomarker in colorectal cancer. Daiichi Sankyo and AstraZeneca's HER2-targeted antibody-drug conjugate Enhertu (trastuzumab deruxtecan) has shown benefit in unresectable or metastatic HER2-expressing colorectal cancer in a Phase II trial. Another study found certain HER2-expressing colorectal cancer patients benefited from the combination of two Genentech anti-HER2 drugs, Herceptin and Perjeta (pertuzumab).
While studies like those back the efficacy of HER2-targeted therapy in HER2-positive colorectal cancer cancers, the data to date have been less supportive of using anti-EGFR treatment in this subset of colorectal cancers.
"There is some consensus that HER2 amplification may be a mechanism of resistance against EGFR inhibitors cetuximab or panitumumab [Amgen's Vectibix]," Lenz said. He added that as the understanding of HER2's role in colorectal cancer has changed, his team wanted to explore the prognostic value of high and low HER2 expression, which occurs more frequently in colorectal cancer, rather than the rarer HER2 amplification.
"[Determining] a clinical predictive and prognostic outcome [of HER2 amplification] with such small numbers is basically not possible," he said.
In the biomarker analysis from the CALGB/SWOG 80405 trial, Lenz and colleagues retrospectively evaluated outcomes for 925 patients in conjunction with an analysis of their HER2 expression levels. Using the NanoString NCounter Analysis System, the researchers conducted RNA gene expression profiling of these patients using a custom 800-gene colorectal cancer-focused panel, which included ERBB2, the gene that encodes for the HER2 protein. The NanoString system totals the number of raw counts that bind with the panel's molecular barcodes to measure RNA expression. Based on the median HER2 expression level of 467 in the cohort, they split the patient population in half and stratified patients into either HER2 high- or low-expressing groups.
Overall, they found that high HER2 expression was generally associated with longer progression-free and overall survival, regardless of the treatment arm.
Across all treatment arms, median progression-free survival in HER2-high expressers was 11.6 months and median overall survival was 32 months. The HER2-low group's median progression-free survival and overall survival were 10 months and 25.3 months, respectively. High HER2 expressers also had improved overall response rates, 63.6 percent in the HER2-high group versus 55.3 percent in the HER2-low group.
HER2 expression appeared to affect treatment outcomes most significantly for patients on Erbitux.
Patients with high HER2 expression had longer median overall survival on Erbitux and chemo, 35 months, compared to Avastin and chemo, 33.6 months. The low HER2 expression group had poorer outcomes in the Erbitux arm compared to the Avastin arm, with median progression-free survival of 9.2 months versus 11 months, respectively. There were no differences in response rate based on HER2 expression and between the individual treatment arms.
In a model exploring the predictive value of this data, the researchers concluded that the response rate, progression-free survival, and overall survival benefit increased as HER2 expression increased, but the benefit plateaued when HER2 expression was higher than the median.
Together, these findings suggested to Lenz and colleagues that high HER2 expression is both predictive and prognostic for treatment outcomes.
The researchers also identified, using Foundation Medicine's next-generation sequencing panel FoundationOne CDx, a small group of patients, 3 percent of the cohort, who had HER2-amplified tumors in this trial. HER2 amplification, which is marked by abnormally high numbers of HER2/ERBB2 gene copies and can indicate a heavier biological dependence on this receptor than in tumors without amplification, usually occurs in only 2 percent of colorectal cancers.
In the HER2-amplified group, the researchers did not identify any significant associations with treatment outcomes and concluded that HER2 amplification was not prognostic for survival or predictive of therapeutic outcomes. Lenz noted that this finding departs from previous research suggesting HER2 amplification is a mechanism of resistance to EGFR inhibition.
"What we found is that HER2 expression level is highly predictive and prognostic, and it's against the assumption that higher HER2 expression is associated with resistance to cetuximab. That [assumption] seems not to be correct," Lenz said, adding that perhaps researchers didn't focus previously on RNA expression levels, but only considered HER2 amplification or immunohistochemistry scores. "But gene expression seems to be much more quantitatively important. It is a new tool to really better understand this marker as predictive and prognostic in the different treatment scenarios."
Because HER2 is becoming a more important biomarker in colorectal cancer, guidelines are changing to reflect the need for testing. Lenz said that US and international guidelines now recommend testing colorectal cancer patients for KRAS, NRAS, and BRAF mutations along with determining microsatellite instability status and HER2 amplification when making treatment recommendations.
However, he noted that HER2 RNA gene expression profiling could help guide treatment even more. The CALGB/SWOG 80405 trial analysis demonstrated that HER2 amplification and HER2 RNA gene expression levels had different effects on treatment outcomes on the EGFR inhibitor Erbitux.
"It's also important as RNA gene expression is not part of the [currently recommended] HER2 testing. [HER2 status] is usually only [gauged via] immunohistochemistry and FISH, but HER2 RNA gene expression may become important in the future to distinguish between patients who benefit the most on bevacizumab and cetuximab," Lenz said. "It's a new target, a new subgroup, and it's not routinely [tested], but it should be."
The researchers wrote that RNA gene expression profiling, as it was done in the CALGB/SWOG 80405 trial, could be a "more effective biomarker for patient selection and stratification" than HER2 amplification. One advantage is that RNA gene expression profiling could also do double duty identifying both patients with high or low HER2 expression and those with HER2 amplification. In this trial, high HER2 expression by RNA profiling was significantly correlated with HER2 amplification by NGS, with all but one patient with a HER2-amplified tumor being classified as HER2 high-expressing.
However, they noted that a limitation of this analysis was that the HER2 expression cutoff was derived from within the cohort. The researchers wrote that future validation studies are needed to determine a standardized HER2 expression cutoff point and to align expression readouts from different RNA expression profiling platforms.
Lenz added that his team has already validated the findings of the CALGB/SWOG 80405 trial in HER2-expressing colorectal cancer in the SCRUM-Japan database, which they hope to present at an upcoming medical meeting.
They have also recently partnered with researchers conducting the Phase III PARADIGM trial, in which Takeda is comparing the EGFR monoclonal antibody Vectibix plus chemo with Avastin plus chemo in colorectal cancer to separately explore the role of HER2 expression on treatment outcomes, according to Lenz.