The rate of agreement between biopsy and resection is generally greater than 90 percent, despite the significant rate of heterogeneity in GEA tumors. Dr. Brownlee speculates that the concordance has to do with the way physicians biopsy tumors. “Frequently they’ll try to biopsy different areas,” she says, “especially if it’s something they can see endoscopically. So our specimens usually have multiple tissue fragments.”
For a biopsy specimen, six to eight fragments should be obtained, though most studies indicate that testing at least three to five biopsies leads to good correlation. As for a resection, “the main considerations are making sure you’ve looked at enough of the tumor to see all the different morphologies, and to select the best blocks of tumor for HER2 testing,” Dr. Brownlee says. The same applies to a metastatic specimen. “Because of tumor heterogeneity, things can vary significantly. Different regions of the tumor, both within the primary mass and metastatic sites, may have different HER2 status.”
When it comes to cytology, an effusion sample is a difficult specimen type. “It lacks tissue architecture and often has less tissue present for evaluation, making it more challenging to interpret,” Dr. Brownlee says. “And to my knowledge, we do not have enough data to compare the efficacy of effusion specimens to intact tissue specimens.” Effusion specimens aren’t common, however, “and usually there are other ways to get tissue, but if it’s the only specimen available to test, we do our best.”
The guideline is clear on the matter of genomic testing: It may be used concurrently or subsequently for clinical decision-making. Although it can be performed concurrently with IHC, “genomic testing is a secondary test for determining HER2 status,” Dr. Brownlee says. “You should not sacrifice tissue and do genomic testing and then not have anything available for IHC/ISH.”
The sensitivity and reliability of alternative genomic testing—whether tissue- or liquid-based—in identifying HER2-positive GEA is relatively low. “Getting amplification data from sequencing is tricky and fraught with error and noise. It’s just not a reliable test,” she says. The guideline reemphasizes that genomic testing should not be used as a standalone test for determining HER2 status.
With the guideline’s sanctioning of concurrent genomic testing, when might such testing be indicated? “If you have adequate tissue and you’re sending a big genomic panel anyway, I think it’s appropriate in that setting,” Dr. Washington says. “That will give us more information in real-life practice settings, especially if we’re doing the IHC concurrently, which many are. If you have to triage your specimen, then that’s a conversation between you and the medical oncologist about the priority for testing.”
Recommendation No. 3 is a distinct departure from the 2016 guideline, which said only advanced GEA patients with unknown or negative HER2 status should undergo repeat testing at time of disease progression. The new recommendation addresses findings from the phase two DESTINY-Gastric01 and Gastric02 trials, in which HER2-positive patients with advanced GEA whose disease progressed on trastuzumab responded to the antibody-drug conjugate trastuzumab deruxtecan. The new recommendation says HER2 assessment may be performed on relapsed/recurrent tumor sample in patients with advanced HER2-positive GEA being considered for subsequent therapy after disease progression. “If they retain HER2 expression, the [antibody-drug] conjugate is indicated and may be beneficial,” Dr. Brownlee says. In patients who progress on or after first-line therapy that includes trastuzumab, acquired clinical resistance to trastuzumab is associated with loss of HER2 positivity in about 29 percent to 60 percent of cases, the guideline says, although the etiology of that loss is not well understood.
“I imagine it’s because tumors evolve over time,” she says. “Going back to the question of tumor heterogeneity, if you have a tumor where some of it is expressing and some of it is not and you give it a targeted therapy, the targeted therapy is going to take care of the part of the tumor that expresses HER2, and the rest of the tumor is going to be resistant and then it can grow and spread. It’s probably a combination of de novo intrinsic factors of the tumor, and then also evolution with the use of therapeutics and the pressure from those therapeutics on tumor evolution.”