Spotlight on ancillary tests in endometrial cancer

Mismatch-repair deficient endometrial cancers tend to be high-grade tumors or endometrioid tumors with mucinous differentiation, Dr. Sayeed said. At VCU, mismatch repair IHC is performed on all biopsies for endometrial carcinoma. “And that is to get ahead of those patients who may not be surgical candidates, so that we can have this information available” for the gynecologic oncologists. At some institutions it may be performed on the hysterectomy specimen only, she said, “which is acceptable as well.”

According to the CAP endometrium biomarker template, “if you have loss of MLH1 and PMS2, you should send for MLH1 [promoter] hypermethylation studies. So we do that at our institution,” she said. Hypermethylated MLH1 confirms a sporadic mutation. “If that is the case, the patients probably don’t need to have genetic counseling to look for Lynch syndrome,” but if the test is performed and it’s negative, patients may need a referral for germline testing.

Any positive reaction in the nuclei of tumor cells is considered intact expression. “So even one tumor cell that has a nuclei stain is considered intact or normal expression,” she said. And it’s common for intact staining to be somewhat patchy, “so I would recommend doing it on the biggest sample of tumor you have so you can account for the heterogeneity.” The interpretation of expression loss in tumor cells should be made only if a positive internal control is seen, she said, such as the nuclei of stromal, inflammatory, or nonneoplastic epithelial cells.

If oncology is requesting mismatch repair deficiency testing repeatedly, Dr. Sayeed said, implementing reflex testing for all endometrial cancers may be helpful. “Don’t be afraid to reach out to your gyn oncology team and ask them, ‘Is there something we can be doing better, or is there anything with biomarker testing that would be helpful for us to set up, or is it something that can be done reflexively?’” When patients are referred to VCU from outside institutions, “one of the most common things I’m encountering when I review their prior pathology is there’s a diagnosis and no biomarkers are done.”

“In this day and age, unless testing is unfeasible due to sampling technique or other technical issues,” Dr. Randall said, “it borders on malpractice not to test an endometrial cancer for mismatch repair deficiency, because that’s how we screen for Lynch. We used to use family history criteria, but that’s not as dependable as mismatch repair IHC.”

The frontline therapy for patients with mismatch-repair deficient endometrial cancer is based on stage, she said. For patients who are chemotherapy candidates, “we’re trying to replace chemotherapy with immunotherapy.” And in many stage three cases, she said, “we’ve stopped giving radiotherapy to give chemo, but for our mismatch-repair deficient patients we may need to give that radiation back. So that’s under active investigation.” Also under investigation is whether immunotherapy should be added to radiation for patients with mismatch-repair deficient early-stage cancers who receive postoperative adjuvant therapy. “We don’t know if those are going to become standard of care,” Dr. Randall said. “It’s still investigational.”

In Fig. 7 is another endometrial tumor (left panel, A), from a patient in her 80s who presented with postmenopausal bleeding. The tumor had a ribbon-like arrangement, with nuclear stratification. “We’re not seeing any glandular formation,” Dr. Sayeed said, but there’s some mitotic activity, and the nucleoli are prominent. “So a high-grade tumor.” Endometrial biomarker testing (right panel, D) revealed a total loss of p53. Mismatch repair deficiency testing showed intact expression. “So given the morphology and p53 status, this was an endometrial serous carcinoma,” she said. These tumors are high grade by definition when the endometrium is the primary, she noted, so grading is not required.

HER2 testing isn’t yet common practice for gynecologic specimens, Dr. Sayeed said. “So what criteria should be applied for HER2 scoring for gyn specimens?” Proposed criteria for ovarian and endometrial cancers have been developed, she said, “but if you use the CAP guidelines, which I do, the endometrial biomarker template still recommends the ASCO breast criteria for HER2.”

HER2 immunostaining (Fig. 8) revealed a lymphovascular invasion (panel A). The HER2 is not perfectly membranous, but it is diffusely positive (C) in the foci of lymphovascular space invasion. More of the tumor is shown in panel B. In panel D, “there’s nice membranous circumferential staining. So even by ASCO criteria this would be diffusely positive for 3+ staining.”

“We don’t often see endometrial cancers that are positive for HER2,” Dr. Sayeed continued. “So we do still send these for FISH, just because we’re starting to perform this at our institution and we want to be sure that our results are confirmed.” In Fig. 9 (specimen B), “you can see the increase in signaling for HER2. So it was positive by FISH for HER2 as well.”

Compared with breast carcinoma HER2 expression by IHC, she said, endometrial serous carcinoma shows higher intratumoral heterogeneity, with at least two degrees of difference in staining intensity involving at least five percent of tumor cells (Buza N. Arch Pathol Lab Med. 2021;​145[6]:687–691; Buza N, et al. Mod Pathol. 2013;26[12]:​1605–1612). “So that’s a good argument for submitting multiple sections of tumor for HER2 IHC interpretation or making sure that you’re sampling well enough to choose the best block for HER2 expression,” she said. “And that would be the block with the most amount of tumor cells.”

The distribution of staining, too, may differ from what’s seen in tumors of the breast, with more basolateral and lateral staining present (Buza N, et al. Arch Pathol Lab Med. 2021;145[6]:687–691; Fader AN, et al. J Clin Oncol. 2018;36[20]:2044–2051). “So the proposed criteria are suggesting that we score HER2 as a 3+ when greater than 30 percent of strong, complete membranous, or basolateral/lateral staining is present—a little different from the ASCO criteria, which looks just at circumferential membranous staining,” she said. These findings, too, support submitting large tumor sections when evaluating HER2 IHC. “At our institution we will submit large sections of tumor when we get requests for HER2 testing,” she said. “And if we get something that looks equivocal or even positive, at the moment we’re sending it for FISH for confirmatory studies, just until we have enough data internally to suggest that we’re using the right criteria.” For now, they’re using the ASCO breast criteria for HER2 scoring, not the proposed criteria.

In metastatic endometrial serous cancers, HER2 expression may differ between the primary tumor and the metastasis, Dr. Sayeed said, noting a 2018 study that found discrepant HER2 scores in 23 percent of cases, with the primary tumor HER2 positive and the metastasis negative (Halle MK, et al. Br J Cancer. 2018;118[3]:378–387). “That’s another reason to sample much of the tumor for HER2 testing, and if needed, make sure you’re testing both the primary tumor and the metastasis,” she said. “We actually had a case at our institution which was the opposite—the primary tumor was HER2 negative but the metastasis was HER2 positive.” So they have been staining both sections of the tumor in metastatic cases, she said, “just to make sure we’re covering all the bases.”

Said Dr. Randall, “We’re starting to morph into those more standardized breast ASCO guidelines.” A clinical trial demonstrated improved progression-free survival in HER2-positive endometrial serous cancers when patients were given trastuzumab as part of their therapy (Fader AN, et al. J Clin Oncol. 2018;36[20]:2044–2051). “And that was true for both initial diagnosis of advanced cancer and recurrent disease,” she said. “So it was true in both treatment settings.”

“We haven’t landed on what the best testing is yet,” she added. “But I will say that all endometrial serous tumors should be tested for HER2 for this reason.” It’s reflex tested at VCU. “You’re not going to find HER2 expression in the nonserous types, even p53 mutant grade three, so we don’t really test those.”

Charna Albert is CAP TODAY associate contributing editor. Drs. Sayeed and Randall’s comments on ovarian and cervical cancer will be published in an upcoming issue.