Gynecologic Oncology

There were 732 evaluable patients with primary, previously untreated, histologically confirmed stage I squamous carcinoma of the cervix with greater than or equal to 3-mm invasion. Of these, 645 had no gross disease beyond the cervix/uterus, had negative paraaortic lymph nodes, and had undergone a radical hysterectomy with pelvic lymphadenectomy. The 3-year disease-free interval (DFIs) for the 545 patients with negative pelvic nodes was 85.6%, and for the 100 with positive pelvic nodes, 74.4%. A large number of pelvic nodes involved with tumor was not correlated with a poorer prognosis; the DFIs were 72.1, 86.4, and 64.6% for one, two, and three or more positive pelvic nodes, respectively. DFI correlated strongly with depth of tumor invasion, both in absolute terms (mm) and infractional thirds. The DFI was 94.6% for less than or equal to 5 mm, 86.0% for 6-10 mm, 75.2% for 11-15 mm, 71.5% for 16-20 mm, and 59.5% greater than or equal to 21 mm. In fractional terms, the DFI was 94.1% for superficial third, 84.5% for middle third, and 73.6% for deep third invasion. With respect to clinical tumor size, the DFIs were 94.8, 88.1, and 67.6% for occult, less than or equal to 3 cm, and greater than 3 cm, respectively. The DFI was 77.0% for those with positive capillary-lymphatic spaces (CLS) and 88.9% for those with negative CLS. Tumor grade and parametrial status correlated with DFI. DFI was not significantly different for age, disease status of the surgical margins, tumor description (e.g., exophytic), quadrant involved with tumor, uterine extension, and keratinizing status of tumor cells. Clinical tumor size, CLS, and depth of tumor invasion were independent prognostic factors.

OBJECTIVE: This study examined both National Comprehensive Cancer Network (NCCN) guideline concordance and steroid taper duration with clinical outcomes in patients with gynecologic malignancies experiencing immune-related AEs. METHODS: This single-institution retrospective cohort included patients with gynecologic malignancies treated with immune checkpoint inhibitor (ICI) from August 2016 to September 2025 who developed grade 3 or 4 immune-related AEs requiring systemic steroids. Clinical and steroid treatment data were abstracted. NCCN guideline concordance and outcomes were assessed. RESULTS: 63 patients contributing 73 steroid tapers were included. Most patients had endometrial (41.3%), cervical (27.0%), or ovarian cancer (23.8%), and pembrolizumab was the most used ICI (74.6%). 46 tapers (63.0%) were NCCN guideline concordant and 27 (37.0%) were NCCN guideline discordant, with discordance most often driven by taper length and premature initiation. 14 (19.2%) were < 28 days, and 59 (80.8%) were ≥ 28 days. NCCN discordant tapers were more likely to require additional supportive medications for irAE management (e.g. secondary steroids) than concordant tapers (81.5% vs 56.2%, p = 0.0300). NCCN concordance was associated with longer time to ICI restart (56 vs 21 days, p = 0.0241). Tapers <28 days were more likely to require additional supportive medications (100% vs 57.6%, p = 0.0027) and steroids (35.7% vs 8.4%, p = 0.0253) than tapers ≥28 days. Tapers ≥28 days were associated with higher rates of ICI discontinuation due to toxicity (59.3% vs 28.6%, p = 0.0381). Rates of hospitalization, ICU admission, and AE resolution did not differ by NCCN concordance or taper duration. CONCLUSIONS: Corticosteroid taper practices varied substantially in patients with gynecologic malignancies experiencing severe irAEs; shorter and guideline-discordant tapers were associated with greater treatment escalation and steroid reinitiation.