Sentinel-Lymph-Node Biopsy Alone or with Lymphadenectomy in Cervical Cancer

Limited data are available on survival outcomes after sentinel-lymph-node biopsy alone as compared with lymphadenectomy in cervical cancer.

In this multicenter, randomized, noninferiority trial, we enrolled patients with cervical cancer that was stage IA1 (with lymphovascular invasion), IA2, IB1, or IIA1 according to 2009 International Federation of Gynecology and Obstetrics criteria. Sentinel-lymph-node biopsy was performed at the time of surgery and was followed by examination of frozen sections. Patients who had negative sentinel lymph nodes were intraoperatively assigned in a 1:1 ratio not to undergo pelvic lymphadenectomy (the biopsy-only group) or to undergo lymphadenectomy (the lymphadenectomy group). All the patients underwent hysterectomy, and adjuvant therapy was provided according to a unified protocol. The primary end point was disease-free survival at 3 years, with a prespecified noninferiority margin of 5 percentage points in the upper limit of the confidence interval for the difference between the lymphadenectomy group and the biopsy-only group. Secondary end points included retroperitoneal nodal recurrence, cancer-specific survival, and surgical complications.

A total of 838 patients underwent randomization: 420 patients were assigned to the biopsy-only group and 418 to the lymphadenectomy group. The median follow-up was 62.8 months. The 3-year disease-free survival was 94.6% in the lymphadenectomy group and 96.9% in the biopsy-only group (difference, -2.3 percentage points; 95% confidence interval [CI], -5.0 to 0.5; P<0.001 for noninferiority); the 3-year cancer-specific survival was 99.2% in the biopsy-only group and 97.8% in the lymphadenectomy group (hazard ratio for death from cancer in competing-risks analysis, 0.37; 95% CI, 0.15 to 0.95). Retroperitoneal nodal recurrences occurred in no patients in the biopsy-only group and in 9 patients (2.2%) in the lymphadenectomy group. The biopsy-only group had a lower incidence of lymphocyst than the lymphadenectomy group (8.3% vs. 22.0%; P<0.001), as well as a lower incidence of lymphedema (5.2% vs. 19.1%; P<0.001), paresthesia (4.0% vs. 8.4%; P = 0.009), and pain (2.6% vs. 7.9%; P = 0.001).

In patients with early-stage cervical cancer, sentinel-lymph-node biopsy alone was noninferior to lymphadenectomy with respect to disease-free survival and was associated with fewer complications. (Funded by Guangzhou Municipal Science and Technology and others; PHENIX ClinicalTrials.gov number, NCT02642471.).