Does Trastuzumab Plus Carbo/Paclitaxel Up PFS in Advanced HER2+ Uterine Carcinoma?

The addition of trastuzumab to the combination of carboplatin and paclitaxel was well tolerated among women with advanced or recurrent uterine serous carcinoma (USC) tumors that overexpress the tyrosine kinase human epidermal growth factor receptor 2 (HER2/neu), according to findings from a randomized phase II trial presented at the 2018 Society of Gynecologic Oncology (SGO) Annual Meeting on Women’s Cancer, held March 24–27 in New Orleans, Louisiana.

The regimen was associated with longer progression-free survival (PFS), but patients in the control group were significantly older than those whose treatment included trastuzumab.

“This encouraging result deserves further investigation to determine the impact of HER2/neu targeted treatment in overall survival in women with advanced or recurrent USC who overexpress HER2/neu,” said Alessandro D. Santin, MD, of the Yale University School of Medicine in New Haven, Connecticut.

USC is a rare, molecularly distinct and aggressive form of endometrial cancer, representing as few as 3% of all endometrial cancers but 40% of all endometrial cancer deaths.

“Up to 70% of women diagnosed with USC have extrauterine spread at the time of diagnosis, and while there is evidence to support benefit of chemotherapy such as carboplatin and paclitaxel, and radiation therapy, the 5-year overall survival (OS) rate remains as low as 41% for all stages,” Dr. Santin said.

Because it is so rare, however, it has been less well-studied than other gynecologic cancers.

USC tends to have a low tumor mutation burden overall, but more than 90% of USC cases harbor TP53 mutations. Gain-of-function mutations in the HER2/neu oncogene occur in 27% of cases, and HER2/neu overexpression is detected using immunohistochemistry testing in 14% to 60% of cases. (“These estimates vary widely in part due to a lack of standardized algorithms for interpretation and scoring of HER2/neu immunostains in endometrial cancer,” Dr. Santin noted.)

HER2/neu overexpression activates the PIK3CA/AKT/mTOR and RAS/RAF/MAPK pathways, and leads to dysregulated gene transcription. Trastuzumab is a humanized monoclonal antibody that targets HER2/neu, inhibiting tumor cell proliferation and triggering apoptosis.

Previous research (the Gynecologic Oncology Group’s GOG181B trial) has suggested that trastuzumab is inactive in HER2/neu-expressing endometrial carcinomas, and in the current study, nearly half of the participants “ultimately did not have tumoral HER2/neu gene amplification,” Dr. Santin said. “Additionally, there was no limit on the number of prior lines of chemotherapy, and much of this cohort had bulky, measurable, heavily pretreated recurrent disease.”

“Since GOG181B, important discoveries have been made that might, at least in part, explain the negative results of that trial,” he said. “Unlike breast cancer, USC is highly heterogeneous in HER2/neu expression. Mutations in the PIK3CA pathway, which may be present in up to 60% of HER2/neu-amplified USCs, have been demonstrated to represent a major mechanism of resistance [to] trastuzumab.”

Dr. Santin and colleagues therefore conducted a randomized phase II trial to examine the effect of adding trastuzumab to carboplatin and paclitaxel treatment for women with HER2/neu-overexpressing USC.

Between 2011 and 2017, a total of 61 patients (58 of whom were ultimately evaluable for efficacy) with advanced or recurrent disease were randomly assigned to a control group receiving carboplatin and paclitaxel for 6 cycles or an experimental group receiving the carboplatin/paclitaxel combination plus intravenous trastuzumab until disease progression or acceptable toxicity.

The patients assigned to the experimental group were significantly younger than those assigned to the control group (median age, 67 years vs. 73 years, respectively; P = .006).

Trastuzumab was associated with a PFS 4.6 months longer than that for the carboplatin and paclitaxel–only control group (12.6 months vs. 8.0 months; hazard ratio, 0.44; 90% CI, 0.26–0.76; P = .005). Clinical benefit and objective response (complete and partial responses) were not statistically significantly different between the study arms.

Adverse event rates were similar between the treatment groups, and no specific toxicity differed significantly between them. One control-group patient died of thromboembolism.

Further investigation is warranted, Dr. Santin concluded. The results will be published in the Journal of Clinical Oncology.