In this large prospective study of women with completely resected UPSC, RT “sandwiched” between paclitaxel/carboplatin chemotherapy is well-tolerated and highly efficacious, when compared to prior studies of single modality adjuvant treatment for UPSC, including in treated patients with advanced stage disease. The toxicities were acceptable and predictably more common following the RT with the vast majority being self-limiting and managed conservatively. These results show additional long term survival and considerably more prospectively tracked efficacy and toxicity data from our prior pilot report.[11
] Patients with both early and advanced stage UPSC who are treated with this “sandwich” protocol have excellent survival, with a 3 year survival probability of 84% in early stage patients and 50% in advanced stage patients. Even better survival was noted among patients who completed all of the prescribed therapy, 85% in early stage patients and 62% in advanced stage patients. These results compare favorably with prior reports in treatments of UPSC.
There is a lack of consensus regarding the standard management of women with UPSC. Most treatments for UPSC appear to involve adjuvant therapy that includes chemotherapy, RT, or both.[4
] However most of the prior reports lack the size, power, or prospective data collection to effectively evaluate the true effect size. To our knowledge, this is the largest prospective trial for this rare tumor, and our survival exceeds most prior reports and obviates the added toxicity of this combination regimen. This protocol was well tolerated with generally self-limiting toxicities and limited dose reductions and dose delays.
Despite the fact that this was a prospective trial, there are limitations. The minor differences in the carboplatin dosing regimens during the pilot phase may raise criticism. The carboplatin dose was optimized during the pilot phase, but all doses were within the therapeutic window for carboplatin. After further optimization during the early part of the pilot and registration phase, the prescribed dose regimen remained consistent. Homogeneity between the groups was confirmed by Cox-regression matched analysis regarding toxic events, recurrence, and survival. Also, this is a single institution trial, with the potential for all biases attributed to any single institution study. However, we believe this is a strength of this study, as well, as most of the investigators were actively working together during this whole trial period and became comfortable with the treatment and managing the toxicity. Also, all subjects who met criteria were approached to register. Only 63% of the early stage patients underwent omental sampling. However, if any of the remaining early stage patients had been misclassified as being understaged as a result of lack of an omentectomy, this would only improve our efficacy results in misclassified early stage patients as they would have been upstaged to stage 4, which would improve the advanced stage survival even more. Additionally, we accrued a limited number (n=14) advanced stage patients to this protocol. This is primarily due to the fact that most patients with metastatic UPSC have measurable disease even after staging surgery. The survival in the patients on this protocol cannot be extrapolated to stage 3&4 UPSC patients with measurable disease after surgery.
While the sequencing of the chemotherapy and radiation therapy in this protocol was 3 cycles prior to RT, then 3 cycles after RT, this might not be the optimal sequence for a “sandwich” therapy. It should be noted that the vast majority of patients (65/67) who received 3 doses of chemotherapy followed by EBRT, were able to complete the remaining three cycles of chemotherapy. The additional three cycles of chemotherapy after RT is well-tolerated. Also, in the subset of 65 patients who completed all cycles of chemotherapy and RT there was a slightly higher survival, though not statistically significant. Another “sandwich” method of sequential adjuvant chemotherapy and radiation was further investigated by Geller et al in advanced stage endometrial cancer.[4
] This was a retrospective analysis of 23 patients with advanced stage III or IV endometrial cancer, including 12 (52%) of the patients with UPSC, who received adjuvant chemotherapy consisting of carboplatin and a taxane, with total number of chemotherapy cycles prescribed ranging from 4-6, with at least 2 cycles post-radiation. Of the 23 patients, 5 progressed and 3 of these patients died during follow-up. The estimated 1, 3, and 5 year PFS was 100%, 80%, and 74% respectively. This is comparable to the study by Obermair and colleagues who used 4 cycles of paclitaxel plus carboplatin followed by pelvic RT.[7
] Further studies on the optimal sequencing in an adjuvant setting in patients with UPSC are warranted. Until other sequences are tested, through, this “sandwich” therapy should be considered and potentially confirmed as a treatment arm in future cooperative group trials.