Abstract
Aim: There is poor evidence regarding effective treatment for recurrent endometrial cancer. We treated patients with recurrent endometrial cancer with docetaxel–cisplatin (DP) therapy as second-line or third-line chemotherapy. We aimed to evaluate the feasibility and efficacy of DP therapy for patients with recurrent endometrial cancer. Patient and Methods: We included 26 patients diagnosed with recurrent endometrial cancer, who underwent DP chemotherapy at our Institution. Docetaxel at 70 mg/m2 and cisplatin at 60 mg/m2 were administered by intravenous injection every 3 weeks. We retrospectively analyzed the clinicopathological factors associated with the response rate (RR) and prognosis. We also analyzed the adverse effects of DP therapy. Results: Median follow-up was 33.8 months and the median number of therapy cycles was six. Grade 3 or 4 adverse effects included leukopenia (66%), neutropenia (81%), anemia (9%), diarrhea (12%), general fatigue (12%), liver dysfunction (4%), peripheral neuropathy (4%), and hyponatremia (4%). RR was 58% and the median progression-free survival (PFS) was 7.5 months. The group with a treatment-free interval of 6 months or more tended to have better PFS than that with less than 6 months (p=0.01). The group with a platinum-free interval of 6 months or more had significantly better PFS than that with less than 6 months (p=0.09). Although the history of taxane usage was not relevant to prognosis, a taxane-free interval of 12 months or more was associated with a tendency for better PFS (p=0.06). Conclusion: DP therapy was fully feasible and demonstrated efficacy for patients with recurrent endometrial cancer.
Endometrial cancer is currently the most common gynecological malignancy (1, 2) in Japan. Although endometrial cancer generally has a relatively good prognosis, the prognosis for advanced cases is reported to be poor (3). The initial therapy for endometrial cancer is surgical treatment, although postoperative chemotherapy or radiotherapy may be added in accordance with the recurrence risk classification on the basis of postoperative pathological diagnosis.
The National Comprehensive Cancer Network (NCCN) (4) and the European Society for Medical Oncology (ESMO) guidelines (5) recommend postoperative radiotherapy for stage I and II endometrial cancer, depending on the risk of recurrence, and chemotherapy for stage III and IV endometrial cancer. In Japan, radiotherapy is rarely performed, whereas chemotherapy is often added to treatment regimens in cases of intermediate to high risk of recurrence (6-9). Thus, most patients with recurrent endometrial cancer have a history of chemotherapy in Japan. Because clinical trials conducted on advanced or recurrent endometrial cancer, such as the Gynecologic Oncology Group (GOG) trial, typically include chemonaïve participants, their findings cannot be extrapolated to patients in Japan. Therefore, no evidence is available regarding therapies for recurrent endometrial cancer previously treated with chemotherapy in Japan. In particular, a suitable chemotherapy regimen remains unclear.
Chemotherapy regimens for advanced and recurrent endometrial cancer include adriamycin plus cisplatin therapy (AP therapy), developed by the GOG 122 trial (10), which significantly prolonged progression-free survival (PFS) and overall survival (OS) in comparison with total abdominal irradiation. Thereafter, paclitaxel with adriamycin plus cisplatin (TAP) therapy was found to prolong PFS and OS significantly more than AP therapy in the GOG 177 trial (11), although the adverse events were also more severe. The recent GOG 209 trial was a noninferiority trial of TAP therapy versus therapy with paclitaxel plus carboplatin (TC); the interim report found that TC therapy was not inferior (12). Watanabe et al. reported that TC therapy is the most common postoperative chemotherapy used in Japan (13). However, although TC therapy has comparatively few adverse events, severe peripheral neuropathy develops as treatment cycles increase (11, 12). Therefore, patients with a history of chemotherapy, particularly TC therapy, have low tolerance to re-exposure.
The Japanese Gynecologic Oncology Group (JGOG) 2041 trial, a randomized phase II trial on chemotherapy for advanced or recurrent endometrial cancer conducted in Japan, found that TC therapy, therapy with docetaxel plus cisplatin (DP), and docetaxel plus carboplatin (DC) had similar therapeutic effects (14).
We retrospectively analyzed the effects and adverse events of DP therapy administered as second-line or later chemotherapy for recurrent endometrial cancer to determine if it is a suitable regimen for patients with recurrent endometrial cancer with a history of chemotherapy.
Patients and Methods
Patients. Participants were 26 patients diagnosed with recurrent endometrial cancer who underwent DP therapy at our Institution. These patients had undergone surgery and postoperative chemotherapy following pathological diagnosis of endometrial cancer between 2002 and 2014. Recurrence was diagnosed pathologically or clinically on the basis of image diagnosis. Twenty-three patients had undergone DP therapy as second-line chemotherapy and three patients as third-line chemotherapy. The median follow-up period was 909 (65-2,725) days. This study was approved by the Keio University School of Medicine Ethical Committee (approval number: 20120236).
Treatment plan. DP therapy was generally administered every 3 weeks, with docetaxel at 70 mg/m2 i.v. at day 1 and cisplatin at 60 mg/m2 i.v. at day 1. Therapy was continued until diagnostic imaging showed that the lesion had disappeared or progressed. Granulocyte colony-stimulating factor was used according to the American Society of Clinical Oncology recommendations (15). Post-treatment for patients who terminated the therapy because of disease progression was not defined.
Evaluation of response and toxicity. Adverse events were assessed with Common Terminology Criteria for Adverse Events (CTCAE) ver. 4.0 (16) on the basis of interviews and blood testing conducted once or more after each cycle. The subsequent cycle started once adverse events were resolved. With regard to hematotoxicity, if patients presented grade 4 leukopenia or neutropenia for more than 7 days, grade 3 or 4 thrombocytopenia, or febrile neutropenia, we considered reducing or withdrawing drugs at the next cycle.
We assessed the overall response rate (RR) of the 26 patients who had received remission induction therapy and had evaluable lesions, in accordance with RECIST ver1.0 criteria (18), and monitored progression-free survival (PFS) and overall survival (OS). Tumors were measured by computed tomography after every two or three cycles. The product of the longest perpendicular diameters was calculated, then the response was assessed using the following criteria: complete response (CR) was defined as the complete disappearance of all known lesions; partial response (PR) was defined as a >30% reduction in the sum of the length of each measurable lesion; progressive disease (PD) was defined as a >20% increase in the sum of the products of all measurable lesions or appearance of any new lesions; stable disease (SD) was any outcome that did not qualify as response or progression.
Clinicopathological characteristics of patients in this study.
Adverse effects after docetaxel–cisplatin therapy of patients in this study.
The treatment-free (TFI), platinum-free (PFI), and taxane-free (TaxFI) intervals were all defined as the period from the final day of administration to the day of diagnosis of recurrence.
Statistical analysis. IBM SPSS Statistics ver. 20 (IBM, Armonk, NY, USA) was used for statistical analysis, using Fisher's exact test, Student's t-test, and Cox's proportional hazard model. Kaplan–Meier curves were used for PFS, and were compared with standard log-rank tests. p-Values of less than 0.05 were considered statistically significant.
Number of cases undergoing docetaxel–cisplatin therapy by recurrence site in this study. PLN: Pelvic lymph node, PAN: para-aortic lymph node, DLN: distant lymph node.
Response rate to docetaxel–cisplatin therapy for recurrent sites. PLN: Pelvic lymph node, PAN: para-aortic lymph node, DLN: distant lymph node.
Progression-free survival (PFS, left) and overall survival (OS, right) following docetaxel–cisplatin therapy by treatment-free interval (TFI, A), platinum-free interval (PFI, B) and taxane-free interval (TaxFI, C). A: Patients with a TFI of ≥6 months tended to have a better PFS than those with a TFI of <6 months. B: Patients with a PFI of ≥6 months also had a significantly better PFS than those with a PFI of <6 months. C: Patients with a TaxFI of ≥12 months tended to have a better OS than those with a TaxFI of <12 months.
Results
Patient characteristics. The median age at initial treatment was 59 (36-72) years, and the median age at start of DP therapy was 62 (38-75) years. The tumor stage, according to the International Federation of Gynecology and Obstetrics 2008 staging system (19), was stage I in 23%, stage II in 4%, stage III in 50%, and stage IV in 23% patients. The histological type was endometrioid adenocarcinoma grade G1 in 19%, G2 in 27%, G3 in 19%, and nonendometrioid type in 35% patients. All patients underwent surgery as initial therapy in most cases (92%) with lymphadenectomy (Table I). Twenty-three patients underwent DP therapy as second-line chemotherapy and three as third-line chemotherapy. Only three patients had previously undergone second-line chemotherapy, with the following regimens: cyclophosphamide-AP-paclitaxel, AP-TC, and AP-docetaxel, respectively.
Univariate and multivariate analyses for progression-free survival (PFS).
Recurrence site prior to DP therapy. The site of recurrence was a single organ in 11 patients and multiple organs in 15. Recurrence was recorded mainly in the lung, peritoneum and pelvic lymph nodes; some patients had recurrence at more than one site (Figure 1).
Adverse effects. The most frequent hematological toxicity-related adverse events of grade G3 or higher were leukopenia (65.4%) and neutropenia (80.7%). No febrile neutropenia was observed (Table II). Grade 3 non-hematological toxicities mainly comprised of diarrhea (11.5%) and general fatigue (11.5%); there were no grade 4 non-hematological toxicities. Dose reduction due to adverse events was required in 8.3% patients, postponement of therapy for 2 weeks or more in 4.2% patients, and discontinuation of therapy in 3.8% patients.
Response to treatment. The median number of treatment cycles was 6 (1-12). The RR (=CR + PR) of DP therapy was 57% and the disease control rate (=CR+PR+stable disease) was 71%. RR by site was high (≥60%) for the vaginal stump, spleen, liver, and lung. By contrast, RR for peritoneal dissemination and lymph node metastasis was low (Figure 2).
Prognosis. The median PFS was 7.5 (4.9-10.2) months and the median OS was 47.9 (27.7-68.0) months. We also investigated factors affecting PFS and OS and found that patients with a TFI of ≥6 months tended to have a better PFS (p=0.050, Figure 3A). Patients with a PFI of ≥6 months also had a significantly better PFS (p=0.034 Figure 3B), and those with a TaxFI of ≥12 months tended to have a better OS (p=0.063, Figure 3C). Furthermore, we investigated factors affecting PFS with a multivariate analysis of age, surgical stage, histological grade, site of recurrence, PFI, and TaxFI and found that only PFI was an independent factor for better PFS (hazard ratio = 0.29) (Table III).
Discussion
The standard treatment for endometrial cancer is surgery with adjuvant therapy such as radiotherapy or chemotherapy, depending on the risk of recurrence and cancer stage. Adjuvant therapy differs greatly between Japan and Western countries. In Japan, chemotherapy is widely used in cases of intermediate to high risk of recurrence, whereas in Western countries, radiotherapy is the main adjuvant therapy and chemotherapy is limited to advanced or recurrent cases. The 5-year OS rate for endometrial cancer in Japan is 86.5%, whereas that of stage III cases (accounting for 20% of all cases) and stage IV cases (accounting for 7% of all cases) is 75% and 32%, respectively (20,21), indicating poor therapeutic outcomes for advanced endometrial cancer. Moreover, response to second-line therapy is also poor. The best response rates are with taxanes, and the GOG has reported paclitaxel to be associated with a 35% response rate in previously untreated women and a 27% response rate in previously treated patients (22). The median survival period reported in clinical trials of advanced cases was approximately 1 year (23).
Treatment for recurrent endometrial cancer depends on the type of recurrence. If the recurrent lesion is localized, surgery and radiotherapy are possible options. However, if recurrence occurs in multiple lesions, chemotherapy is typically selected. Most of the evidence in favor of chemotherapy for recurrent endometrial cancer comes from clinical trials for both advanced cancer and recurrent cancer cases.
In a phase II study for advanced or recurrent endometrial cancer, the RR to single-agent docetaxel therapy for endometrial cancer was 31%. CR and PR were 0% and 23%, respectively, in patients with a history of chemotherapy, and 5% and 32%, respectively, in chemonaïve patients (24). In another phase II trial of recurrent endometrial cancer, RR to weekly docetaxel therapy was only 8%, as all patients (n=48) in the trial had a history of chemotherapy (25). These results are comparable with the RR to doxorubicin, a key drug for endometrial cancer, and paclitaxel, which has recently reached widespread use. On the other hand, RR to chemotherapy in patients with a history of chemotherapy is generally lower.
In the GOG 107 phase III trial of single-agent doxorubicin therapy and DP therapy for advanced and recurrent endometrial cancer, RR and PFS were significantly better in cases administered the combination therapy (25). Moreover, combination chemotherapy regimens such as TC resulted in response rates ranging from 38% to 76% (21-23). These trials obtained better outcomes with combination therapy than single-agent therapy; therefore, most of subsequent clinical trials have compared combination therapies. In Japan, the JGOG 2041 randomized phase II trial investigated DP, DC, and TC therapies for advanced and recurrent endometrial cancer (17). In this trial, 37% patients (n=29) had a history of chemotherapy, and RR to therapy was 51.7%, 48.3%, and 60.0%, respectively, indicating no significant difference between therapies in terms of RR and survival. However, these clinical trials were either limited to chemonaïve patients, or only included a small number of patients with a history of chemotherapy. Therefore, there are few reports of clinical trials in which chemotherapy is used as a second-line or later therapy in patients with recurrent endometrial cancer. In contrast, many recurrent cases in Japan have a history of chemotherapy as adjuvant therapy. According to Watanabe et al., TC and AP therapies account for 60% and 24%, respectively, of adjuvant chemotherapy for endometrial cancer in retrospective studies in Japan (13). The same results apply to the situation of adjuvant therapy at our Institution and all the patients included in this study had a history of chemotherapy.
The reasons for choosing DP therapy were its anticipated efficacy for endometrial cancer, as described above, as well as the fact that only mild peripheral neuropathy occurs as an adverse event. This makes it useful for patients who have undergone AP therapy as prior chemotherapy but cannot undergo AP therapy again because of cardiotoxicity, as well as for patients who have undergone TC therapy but who find repeated treatment difficult because of a markedly reduced quality of life as a result of peripheral neuropathy.
The RR was 57% in the present study, similar to the RR of prior clinical trials of DP therapy. Moreover, the RR of parenchymatous organs was better, whereas RR of lymph node metastases was lower. The latter finding is expected as minimal effect is generally obtained in cases where the primary metastasis is found in the lymph nodes. Furthermore, we observed neutropenia of grade G3 or worse in over 80% of patients, but therapy had to be discontinued because of adverse events in only 3.8% patients. We were able to administer a median of six cycles, rendering this therapy feasible in patients with a history of chemotherapy. Of particular note, a low proportion (4.8%) of sensory peripheral neuropathy of grade G3 or worse was found in this study and the quality of life of the patients was not markedly reduced in the course of DP therapy. However, this was a retrospective study in which adverse events were assessed on the basis of data from the medical records; thus, the assessment was undeniably limited.
The median PFS was 7.5 months, whereas the median OS reached 47.9 months in the present study. This long median OS is very different from the median OS for patients with ovarian cancer, and is likely because of the specific effects of radiotherapy and hormonal therapy in recurrent endometrial cancer, which saved the patients.
We also investigated clinicopathological and prognostic factors in the present study by univariate and multivariate analysis and found that PFI was a prognostic factor. This is similar to ovarian cancer. A previous report found the effect of a TFI similar to that of ovarian cancer in endometrial cancer on second-line and later chemotherapy (29). Another report concluded that OS was good in patients with a PFI ≥6 months, and an RR of 41.7% was obtained in such patients (30). The present study also found that patients with a PFI of ≥6 months had a significantly better prognosis, which is consistent with the results of previous reports.
In summary, DP therapy is feasible and the treatment demonstrates good efficacy for recurrent endometrial cancer, particularly in patients with a PFI ≥6 months pre-chemotherapy. However, because this was a retrospective study, the optimal regimen should be investigated in a phase II/III prospective study comparing several regimens including DP therapy for patients with a history of adjuvant chemotherapy.
Acknowledgements
The Authors thank Ms. Keiko Abe and Ms. Tomomi Noda for their secretarial assistance.
This study was partially supported by the grant of Japan Ministry of Education, Culture, Sports, Science and Technology, Basic Research Foundation-C (no. 25462612).
Footnotes
Disclosure
The Authors declare no conflict of interest in regard to this study.
- Received December 22, 2015.
- Revision received February 29, 2016.
- Accepted March 9, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved








