By Fred Gebhart
SAN FRANCISCO -- March 17, 2010 -- Women treated for advanced endometrial cancer who have 1 positive lymph node or cervical stromal involvement may do better on systemic chemotherapy than radiation, according to a new study from the Gynecologic Oncology Group (GOG) presented here on March 15 at the Society of Gynecologic Oncologists (SGO) 41st Annual Meeting on Women’s Cancer.
A post hoc analysis of the GOG-122 trial found that doxorubicin plus cisplatin (AP) is associated with better survival than whole-abdomen irradiation (WAI) in women with cervical stromal involvement or a positive pelvic/para-aortic lymph node. Stromal involvement and positive nodes are also associated with a greater risk of progression and death.
“GOG-122 changed the landscape,” said Krishnansu Tewari, MD, University of California Irvine, Orange, California. “There is now a lot of consideration given to eliminating radiotherapy in favour of systemic chemotherapy for advanced disease. Our goal was to determine whether the treatment effect seen in GOG-122 is sustained in the setting of nodal factors or cervical stromal involvement.”
The original GOG-122 trial compared WAI and AP in women with advanced (International Federation of Gynecology and Obstetrics stage III/IV) endometrial cancer after surgery. Although systemic chemotherapy produced higher rates of serious adverse events than radiotherapy, it also produced longer progression-free survival (PFS; 50% at 60 months vs 38%) and better overall survival, with a hazard ratio of 0.71 (P < .01).
A retrospective analysis of GOG-122 found that the original findings in favour of chemotherapy continue in nodal subgroups. Hazard ratios for both PFS and overall survival for women with lymph-vascular space involvement (LVSI) who received chemotherapy were all <1. For PFS, hazard ratios ranged from 0.671 for LVSI not recorded to 0.663 for LVSI negative and 0.792 for LVSI positive. For overall survival, the hazard ratios were 0.712 for LVSI not recorded, 0.946 for LVSI negative, and 0.730 for LVSI positive.
The results for cervical stromal involvement were similar. For PFS, the hazard ratios were 0.672 for no stromal involvement and 0.836 for positive stromal involvement. For overall survival, the hazard ratios were 0.679 for no stromal involvement and 0.852 for positive stromal involvement.
The results for positive lymph node involvement were mixed. In patients with >1 positive node, the hazard ratio for PFS was 1.273 whereas it was 0.959 for patients with only 1 positive node. The hazard ratio for overall survival was 1.049 for >1 positive node and 0.731 for 1 positive node.
“Before 2006, we would have irradiated these women,” Dr. Tewari said. “Do we really want to subject them to radiation? Combining these 3 prognostic factors -- LVSI, cervical stroma, and pelvic nodes -- chemotherapy is favoured for PFS and overall survival.”
[Presentation title: The Prognostic Impact of Number of Positive Lymph Nodes and Cervical Stromal Involvement Among Patients With Advanced Endometrial Cancer Treated With Whole-Abdomen Irradiation Versus Systemic Chemotherapy: A Gynecologic Oncology Group Study. Abstract 8]
