More studies are needed to refine treatment strategies in post-trastuzumab advanced GEA patients with and without persistent HER2 positivity. “The main thing is learning more about the natural history of gastroesophageal carcinoma in the context of the usage of these therapeutic agents,” Dr. Brownlee says. “What are the common resistance mechanisms? Are these things that we need to test for in addition to HER2 status? What is the role of the tumor microenvironment in resistance to therapy? We need to know more about the disease biology,” she says.
The guideline does not make changes or add recommendations to the scoring of gastric HER2 IHC or ISH (a patient with a score of 0 or 1+ is negative, 2+ is equivocal and should be followed by ISH, and 3+ is positive). Use of the terminology “HER2-low” is not recommended for reporting at this time. “We haven’t been doing HER2 as long in gastric cancer,” Dr. Brownlee says of the lack of data, unlike in breast cancer.
Still, pathologists may have questions about HER2-low at this disease site, Dr. Washington acknowledges. “We want to get it right and we want to give our clinician colleagues the information they need concerning biomarkers. HER2-low is an evolving field, but pathologists can take comfort in that they don’t need to change the way they’re scoring HER2,” she says.
“The HER2-low story will play itself out over the next few years.”
The incidence of gastroenteropancreatic neuroendocrine tumors (GEP-NETs) has been increasing steadily, especially in the older population.
Thanks to greater use of endoscopy and advanced imaging, physicians are identifying more of these patients, says Chanjuan Shi, MD, PhD, co-chair of the new guideline on GEP-NETs developed by the CAP in collaboration with the North American Neuroendocrine Tumor Society and the European Neuroendocrine Tumor Society. “We get more and more new patients with this disease now,” says Dr. Shi, professor of pathology at Duke University School of Medicine.
Prevalence has risen even more than incidence, given the disease’s relatively good prognosis. “We need to keep treating and grading these patients,” says Dr. Shi, who will present the new guideline at the USCAP meeting.
Tumor grade is a key prognostic factor for GEP-NETs, Dr. Shi says, and it determines much about the patient’s treatment and surveillance. The World Health Organization recommendations for grading NETs are well established: The Ki67 proliferative index should be evaluated in hot spots with the highest labeling, with at least 500 tumor cells counted, and mitotic rates should be assessed in the most mitotically active areas.
“These are all the guidelines we have,” she says, noting they leave important points unresolved. For example, can Ki67 be evaluated in small specimens? Second, if the pathologist has multiple specimens from the same patient, should all be tested or just some, and if so, which ones? Questions about methodology remain as well. Because discordance between Ki67 grade and mitotic count is frequent, and when there is discordance, the higher grade—almost always the Ki67—should be assigned as the tumor grade, can mitotic count be dispensed with? In addition, the WHO doesn’t provide specific guidance on which method to employ to calculate Ki67.
When the new guideline is published, pathologists and their clinical colleagues will have a better blueprint for navigating these tumors. The guideline also may come in handy for stakeholders in the software space who are developing methods to automate digital counting of Ki67, she says.
The guideline lays to rest the question of small samples, with a recommendation that says pathologists should perform Ki67 grading on fine-needle aspirates/fine-needle biopsies or endoscopic biopsies. “There are multiple reasons for that,” Dr. Shi says. For one, the tumor grade helps guide provision of immediate therapy, whether surgical or medical. “The second thing is this might be the patient’s only specimen,” and thus the only way to know the tumor grade. Moreover, studies suggest good concordance rates between small samples and resection specimens, although rates of discordance are not trivial, especially for grade 2/3 NETs. There’s always the possibility, too, that the tumor isn’t a neuroendocrine tumor at all. “Before we do a biopsy, we don’t know if it’s a neuroendocrine tumor that’s indolent or if it’s an aggressive cancer,” she says. A high Ki67 may indicate a different diagnosis, the guideline says, such as GEP-neuroendocrine carcinoma or acinar cell carcinoma in the pancreas.
Tumor grade heterogeneity is the typical culprit for the discordance between small samples and resection specimens. With small samples, “you can get a cold spot rather than a hot spot,” Dr. Shi says. “That’s the main mechanism that causes discordance.” Undergrading is more likely than upgrading in that scenario, which the guideline’s authors address by recommending that pathologists include with the report a note about the potential for misgrading on small specimens. The guideline also suggests that Ki67 evaluation not be performed in specimens with fewer than 100 tumor cells, unless there is suspicion for a neuroendocrine carcinoma or another more aggressive neoplasm.