PracticeUpdate Oncology May 2019

CONFERENCE COVERAGE 24

Tumor Grade/Size, Lymphovascular Invasion Predict Lymph Node Involvement in Early Endometrial Cancer Findings may help identify patients at low enough R isk factors that include lymphovascu- lar invasion as well as tumor grade and size are powerful independ-

To try to identify these patients, the inves- tigators collected data on all 429,084 patients with endometrial cancer from the study years 2004 to 2015 who were included in the National Cancer Data- base. From this group, they analyzed data on the 4186 patients with adenocarcinoma histology, American Joint Committee on Cancer pathologic tumor stage T2, at least 10 lymph nodes sampled at the time of sur- gery, as well as complete data regarding lymphovascular space invasion and tumor grade. The investigators calculated the proportion of patients with pelvic or paraaortic lymph node involvement as a function of multi- ple demographic, clinical, and pathologic factors. The findings were presented by Dr. Kamrava’s resident, Eric Anderson, MD. Univariate analysis revealed that race (P = .003), tumor grade (P < .001), lympho- vascular invasion (P < .001), tumor size (P < .001), and number of lymph nodes examined (P < .01) were all predictors of lymph node involvement. Multivariate analysis conducted in 3428 patients revealed predictors of lymph node involvement to include black versus white race (odds ratio 1.450, 95% confidence interval 1.098–1.908, P = .008), tumor grade 3 versus 1 (OR 1.441, 95% CI 0.996–1.564, P = .08), the presence of lymphovascular invasion (OR 4.845, 95% CI 4.090–5.753, P < .01), tumor size (OR 1.005, 95% CI 1.002–1.007, P < .01 for each incremental increase of 1 mm), and number of regional lymph nodes examined (OR 1.010, 95% CI 1.002–1.018, P = .016). Dr. Kamrava acknowledged that he was surprised that race emerged as an inde- pendent predictor. “There are a lot of data that, stage for stage, African Ameri- can women tend to do worse, but no one knows why. It looks like, on average, that they had a higher risk of having nodes involved.” “If you had no lymphovascular invasion and a grade 1 tumor, then you had less than 10% of risk of lymph node involvement,” he continued, “but if you had lymphovas- cular invasion and grade 3 disease, that was almost a 50% risk of lymph node involvement.” These data can be used to help guide the discussion of appropriate therapy among physicians and their patients, he con- cluded. www.practiceupdate.com/c/81086

ent predictors of lymph node involvement among women with stage II endometrial adenocarcinoma, according to a poster presented at ACRO 2019. “In endometrial cancer, the standard treat- ment paradigm is to do surgery first and, based on pathological findings, some- times do additional treatment, either radiation alone or in combination with chemotherapy,” senior study investigator Mitchell Kamrava, MD, of Cedars Sinai Med- ical Center in Los Angeles told Elsevier’s PracticeUpdate . “When thinking about radiation, there are two choices we usually make,” he contin- ued. “One is to just treat the vaginal cuff [using] brachytherapy, which can be done relatively quickly, say, within three treat- ments. But, if you feel that the lymph nodes may [harbor] microscopic disease, then you cover that area as well using external beam radiation therapy. That is usually given over 5 weeks. It’s a lot more time, and it exposes more normal tissue [to radiation], so it can carry more side effects.” A recent pooled analysis of just over 100 patients suggested that rates of recurrence are low in women with stage II endometrial cancer with a low risk profile who received radiotherapy to the vaginal cuff alone. “This suggested there may be a subset of patients in whom you do not need to treat the pelvis," said Dr. Kamrava.

risk to consider skipping radiotherapy of the entire pelvic area.

" If you had no lymphovascular invasion and a grade 1 tumor, then you had less than 10% of risk of lymph node involvement, but if you had lymphovascular invasion and grade 3 disease, that was almost a 50% risk of lymph node involvement. "

PRACTICEUPDATE ONCOLOGY