NEW YORK (GenomeWeb) – Investigators are making a case to colleagues in the precision oncology field that their findings in a recent study — showing colorectal cancer patients with HER2 amplification have poor responses to EGFR targeted therapy — should be taken into consideration in clinical practice and could help improve patient care.
Published in late January in JCO Precision Oncology, the study concluded that HER2 amplification in RAS and RAF wild-type colorectal cancer is predictive of a lack of response to anti-EGFR drugs, enough so that these patients should potentially be considered for alternative clinical trials before anti-EGFR treatment.
Although the data is retrospective and relatively limited so far, first author Kanwal Raghav, a member of the gastrointestinal medical oncology division at MD Anderson Cancer Center, said that he and his colleagues consider the current study results strong enough to inform clinical practice at their institution, where they now funnel colorectal cancer patients with HER2 amplification into various HER2-targeted clinical trials instead of to second-line anti-EGFR therapy as a matter of course.
That there is a connection, and potential interference, between HER2 and EGFR is not new, Raghav said in an interview this week. "It's very logical, because EGFR is basically HER1, so this is the same family of receptors … and they all work to the same downstream pathways, so it makes sense that if you blocked EGFR and you have an active HER2 then that same pathway can still be activated."
"It's the same story as KRAS mutations," he added. "If you have another way of activating the pathway then what's the point of blocking EGFR."
But despite this being known, the relative rarity of HER2 amplification in colorectal cancers had made it difficult to collect appropriate data actually showing that patient response suffers.
In their JCO study, Raghav and his colleagues retrospectively studied a group of RAS/BRAF wild-type CRC patients from two distinct cohorts: 98 patients in which HER2 amplification was tested using dual in situ hybridization, and another 70 patients who had next-gen sequencing that included HER2.
The team found that median progression-free survival for patients in the first cohort was significantly shorter for those that had HER2 amplification compared with those that didn't (2.8 versus 8.1 months, respectively).
The same was seen in the second cohort with a median PFS for HER2 amplified individuals of 2.8 months versus 9.3 months. And in multivariable analyses, HER2 amplification was clearly a single independent predictor of poor PFS on anti-EGFR therapy for both groups.Â
Over the last several years, the therapeutic potential of anti-HER2 drugs in tumor types other than breast cancer (for which they were first developed) has become an increasing area of interest. And several trials, including MyPathway in the US, and HERACLES in Europe, have since shown that that HER2 is indeed a biomarker of response to these drugs in colorectal tumors.
According to Raghav, this positive predictive value of HER2 amplification, and the increasing availability of clinical trials for both existing and new HER2 drugs, is part of what underscores the utility of his group's conclusions regarding the negative predictive nature of HER2 in regard to EGFR drugs, despite limitations of the current study.
Unlike other biomarkers of non-response to anti-EGFR therapy like KRAS, which tells a clinician that a patient won't respond to the drug in question, but "offers no other options," a HER2 amplification finding not only suggests a patient maybe shouldn't get an EGFR-targeted drug, but also that they really should get on a HER2 medication.
"If you look at the HERACLES study that was done in Europe and the MyPathway study that was done in the US, those were dual anti-HER2 therapies done on two different continents on patients who had gotten every line of treatment that you can think of for colon cancer. And despite that, they had response rates of 35 to 50 percent, which is [better than] the response rate of anti-EGFR in a selected group, period," Raghav said.
"In second line, which is the commonest [use of anti-EGFR drugs] today, response is about 30 percent." Not to mention that HER2 drugs are, on average, safer and more pleasant than EGFR inhibitors, he added.
Finally, Raghav argued, there is far from a dearth of trials available, many of which are open in multiple centers in the United States.
The main objection he feels other clinicians may have to the conclusions of his team's study is that despite clinial trial ability there will still be cases where there isn't a trial option for whatever reason, and in those cases, shouldn't the patients still use an EGRF inhibitor?
"My answer is, you know the current evidence would say that yes, maybe if you are stuck in the last resort … you can try anti-EGFR if you really want," he said. "But you should at least have that discussion with the patient that you know there is a benefit and risk here."
While EGFR amplification is more rare in the overall colon cancer population, it rises up to at least 8 percent in the subset of RAS and RAF wild-type individuals who are eligible for EGFR drugs. And at that point the question of benefit becomes more significant Raghav argued.
Based on the group's study results and the data from trials like MyPathway, he reiterated that there is real reason to believe that it better serves patients to try to refer them to HER2 drug trials than to treat them with a therapy "where it is very hard to establish what would happen."
Raghav added that current guidelines don't yet deal very specifically with HER2 testing in colorectal cancer. "NCCN has been debating this issue for quite some time … but they haven't endorsed it as yet," he said. "But hopefully with more and more data that's generated we can get there."
Regardless of guidelines, two things may smooth the way for integrating HER2 status into the care of colorectal cancer patients, he added. For one, the field is seeing increasing use of broad sequencing panels in patients with advanced solid tumors, and these panels include HER2, which means that clinicians don't need to separately advocate for investigating this biomarker in their CRC patients.
But even in lieu of this type of precision medicine approach, getting a readout on HER2 is relatively easy. "The easiest way to check this is [immunohistochemistry] and ISH, which is a readily available procedure. You can just send it to your pathologist because all those people … even for a low volume pathologist, the most common testing is breast cancer HER2," Raghav said.
"The reagents are the same, the procedure is the same, the reading is the same, so as long as you have a pathologist who has ever read a HER2 for breast cancer or for gastric cancer they can do that reading for colon cancer," he added.