In this retrospective cohort study, we analyzed data from 20 patients with EC treated with LEN/PEM between January 2022 and March 2023 at IMUH, Japan. It is important to evaluate the safety and effectiveness of LEN/PEM treatment in real-world clinical settings because it entails treatment of patients with diverse clinical characteristics that may extend beyond the selection criteria of clinical trials. Given that UCS, which comprised 30% of the study cohort, was not included in the Study 309/KEYNOTE-775, this study provides important data to physicians in this field. Moreover, our study showed that patients with p53abn had poor prognoses even after treatment with LEN/PEM therapy, based on histomorphology and immunostaining results assessed according to the ProMisE classification. Overall, the ProMisE classification system is an important tool for improving personalized treatment of patients with EC by allowing for more accurate risk stratification and tailored therapeutic decision-making. Although the MMR and p53 status can be easily detected via immunohistochemical staining, there is currently no standard method for identifying POLE mutations, which are generally identified using next-generation or Sanger sequencing. Herein, we did not perform mutational analysis of the POLE gene; however, we did not find any cases with histomorphology suggestive of the presence of POLE, as reported by Van Gool et al. [20]. On the other hand, the safety profile was similar to that reported in previous clinical trials, corroborating the results of recent studies in a Japanese cohort, and no new safety signals were discovered.
The diagnosis of POLE using the ProMisE classification is difficult at the routine clinical level. Immunohistochemical examination of other molecular diagnostic subtypes proposed by the TCGA, such as p53 for the CNH type and MMR-related proteins for the MSI type, is recommended. On the other hand, molecular biological methods are the only available diagnostic tools for identifying POLE gene mutations, and no surrogate markers have been identified. In this study, we tested the identification of POLE using a simple combination of morphological and immunohistochemical characteristics (tumor type, grade, peritumoral lymphocytes, MLH1, and p53 expression) based on the study by Van Gool et al. [20] They reported that this approach helped prescreen for POLE exonuclease domain mutations in ECs, with an estimated sensitivity of 80% and specificity of 88% [20]. POLE type EC reportedly accounts for approximately 10% of all ECs [21,22]. However, the prognosis is extremely good [11,22], and many cases are diagnosed at an early stage; [22] therefore, there are very few reports of recurrent cases [23]. Further investigations are needed to efficiently detect rare POLE type in patients with recurrent EC.
The ORR for the 20 patients was 60.0%, and 2 of 6 patients with UCS exhibited a response. An initial small retrospective case series by Hunt et al. [24] that included 7 UCS patients with UCS treated with LEN/PEM found no partial or complete response. However, that study did not include patients who had received only one line of prior therapy. The results of a larger scale phase II trial of LEN/PEM therapy for uterine and ovarian carcinosarcomas (NCT05147558), which is currently underway to confirm its efficacy, are awaited. Previous reviews have shown that the majority (73.9%) of UCS lesions are categorized into the serous-like p53-abn group, whereas NSMP is relatively rare (13.5%) [25]. The prognosis of the NSMP group is known to be extremely diverse: UCS and p53abn have a poor prognosis and low-grade EC has a good-to-intermediate prognosis [26]. Since both patients with UCS who responded to LEN/PEM in this study showed NSMP, aggressive LEN/PEM therapy may be worthwhile in this population. In contrast, p53abn is not only associated with poor prognosis in UCS but also in the entire spectrum of EC [27]. This finding is supported by our study, which showed that the PFS after LEN/PEM treatment than was significantly shorter in the p53abn group than that in the MMRd/NSMP group. The observed PFS was extremely poor at 3.2 months (95% CI: 2.3-4.2). Improving the outcomes in this group necessitates considering the differentiation between platinum combination therapies and administering treatment at earlier stages, such as in the ENGOT-en9/LEAP-001 study [28] as well as developing novel therapies based on novel biomarkers, such as ERBB2 (HER2) [29,30]. More recently, trastuzumab deruxtecan, an antibody-drug conjugate targeting HER2, was shown to be highly effective against UCS. [31] Although the current guidelines do not include molecular subtyping in clinical decision algorithms, the appropriate use of these drugs based on biomarkers ought to be clarified in the future.
As observed in other studies using VEGF-receptor tyrosine kinase inhibitors, the principal side effects of LEN were hypertension, diarrhea, fatigue, nausea, and vomiting [32]. The efficacy and safety of LEN monotherapy as second-line treatment for recurrent EC have been reported. Hypertension was the most frequent cause of treatment discontinuation in all grades and grade 3 (49% and 31%, respectively) in the study. [6] In this study, AEs associated with LEN/PEM were common and comparable to those encountered in Study 309/KEYNOTE-775 and Study 111/KEYNOTE146. It is noteworthy that all an imaging response was achieved in four patients with clinically evident irAEs. This finding is consistent with previous studies on non-small-cell lung cancer. [33,34]. However, other studies suggest that severe irAEs (grade 3 or higher) may result in a poor prognosis compared to mild irAEs. Additionally, studies have reported that high-dose corticosteroid administration [35] or discontinuation of immune checkpoint inhibitors [36] significantly worsen PFS and OS. Ultimately, these findings emphasize the importance of the early detection of irAEs and appropriate early intervention. Anti-angiogenic therapy with immune checkpoint inhibitors is now an integral part of the management of gynecological cancers. The side-effects of anti-angiogenic drugs and immune checkpoint inhibitors need to be distinguished as their therapeutic approaches are quite distinct, and the management of these AEs is key to treatment success.
Our study has several limitations. First, it was a single-institutional retrospective analysis that extracted data from physician descriptions and laboratory information stored in electronic medical records. Unlike prospective clinical trials, the sample size was not defined in advance, the number of cases was small, and the timing of visits and tests was not defined, which could potentially result in delayed detection of efficacy or disease progression. In addition, minor AEs that were not of concern to the physician in charge could not be identified, leading to a potentially lower incidence rate. Second, although this study evaluated treatment recurrence, no quality-of-life survey was conducted. Finally, this was a single-arm study with no control group. Therefore, it is not possible to determine whether the presence of p53abn in patients treated with LEN/PEM is a prognosticator or predictor of efficacy. Nevertheless, this study provided important data on the effectiveness and safety profile of LEN/PEM, which will be valuable for physicians treating patients with EC.
In conclusion, this study has significant clinical implications for the treatment of recurrent EC. The novel ProMisE molecular subtype, p53abn, has a poor prognosis, irrespective of histology, even with LEN/PEM therapy, and novel therapeutic strategies are needed to combat these tumors.