- Email: [email protected]
Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy
YGYNO-977069; No. of pages: 6; 4C: Gynecologic Oncology xxx (2018) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy Lucia Tortorella a, Carrie L. Langstraat a,b, Amy L. Weaver c, Michaela E. McGree c, Jamie N. Bakkum-Gamez a, Sean C. Dowdy a, William A. Cliby a, Gary L. Keeney d, Mark E. Sherman e, Saravut J. Weroha b, Andrea Mariani a, Karl C. Podratz a,⁎ a
Division of Gynecologic Surgery, Mayo Clinic, Rochester, MN, USA Division of Medical Oncology, Mayo Clinic, Rochester, MN, USA c Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA d Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA e Department of Health Sciences Research, Mayo Clinic, Jacksonville, FL, USA b
H I G H L I G H T S • Most uterine serous carcinomas fail to respond to platinum-based chemotherapy. • Critical lag times precede declaration of platinum refractory microscopic disease. • Truncated survival/systemic failures compel a search for more effective therapy.
a r t i c l e
i n f o
Article history: Received 27 December 2017 Received in revised form 20 February 2018 Accepted 28 February 2018 Available online xxxx Keywords: Platinum-based chemotherapy Therapeutic efficacy Uterine serous carcinoma
a b s t r a c t Objective. Two randomized trials failed to demonstrate efficacy of platinum-based chemotherapy (PbCT) for uterine serous carcinoma (USC). Our objective was to reassess the value of PbCT for patients with microscopic residuum (R0). Methods. Progression-free survival (PFS) after surgery was analyzed for 409 patients and correlated with adjuvant therapies: vaginal brachytherapy (VBRT), external beam radiotherapy (EBRT), PbCT, or combinations. Results. The estimated 5-year PFS for stage I (n = 209) USC was 65.1% for observation only; 90.7%, VBRT only; and 91.1%, PbCT ± VBRT (85% received VBRT); VBRT significantly (P = .004) impacted PFS, but the added value of PbCT remains uncertain. Of 58 stage IIIC, PbCT-treated patients (±EBRT), 5-year PFS was 33.9%; most failures had a vascular disseminated component. Median PFS for 72 stage IV, PbCT-treated patients was 18.6 months for R0; 8.0, R1 ≤ 1 cm residual disease; and 4.6, R2 N 1 cm (P = .008). The progression rate (PR) during 1 to 2 year followup for R0 was similar to PR during 0–1 year follow-up for R1 (P = .31), suggesting recurrences in patients with R0 disease before 2 years are likely platinum resistant. PRs during follow-up were nearly identical for R0 ≥ 2 years and R1 ≥ 1 year (P = .95), presumably showing limited numbers of platinum-sensitive tumors. Conclusions. A comparison of PR for patients treated with PbCT for stage IV R0 and R1 disease suggested that a 1-year lag interval precedes clinical recognition of PbCT refractory/resistant R0 disease. Most patients treated with PbCT who had microscopic residuum had recurrences within 2 years (across stages), emphasizing the need for more effective therapy. © 2018 Published by Elsevier Inc.
Abbreviations: CT, chemotherapy; EBRT, external beam radiotherapy; EEC, endometrial endometrioid carcinoma; HGSOC, high-grade serous ovarian cancer; HR, hazard ratio; IQR, interquartile range; OS, overall survival; PbCT, platinum-based chemotherapy; PFS, progression-free survival; PR, progression rate; R0, no macroscopic residual disease; R1, residual disease ≤1 cm; R2, residual disease N1 cm; RT, radiotherapy; USC, uterine serous carcinoma; VBRT, vaginal brachytherapy. ⁎ Corresponding author at: Division of Gynecologic Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905, USA. E-mail address: [email protected] (K.C. Podratz).
1. Introduction The current therapeutic algorithm for advanced-stage, high-grade uterine epithelial cancers generally includes surgical staging, cytoreduction, and administration of adjuvant platinum-based chemotherapy (PbCT) [1]. These principles have been extrapolated from the management of advanced-stage, high-grade serous ovarian cancer (HGSOC), for which the response rate to PbCT is 80%, and median
https://doi.org/10.1016/j.ygyno.2018.02.022 0090-8258/© 2018 Published by Elsevier Inc.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
2
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
overall survival (OS) rates exceed 5 years in patients whose tumors are completely cytoreduced [2]. This experiential model has been advocated for both advanced endometrial endometrioid carcinoma (EEC) and uterine serous carcinoma (USC). However, in the protocol based on Gynecologic Oncology Group 122, a clinical trial that randomized patients with advanced and recurrent endometrial cancer to either abdominal irradiation or PbCT, a survival advantage was readily shown with PbCT for EEC (progression-free survival [PFS] hazard ratio [HR], 0.687; OS HR, 0.482) but not for USC (PFS HR, 0.909; OS HR, 1.025) [3]. Similarly, a prospective clinical trial of pelvic radiotherapy (RT) alone vs RT plus PbCT for patients with stage I/II endometrial cancer showed a survival advantage with added PbCT for patients with EEC but not for those with USC or clear cell carcinomas [4]. Nevertheless, the increasing use of PbCT for treating USC at our institution and others, even for stage I patients, presupposes an evolving standard of care with uncertain justification [1]. This transition in clinical practice patterns is based on numerous retrospective studies, reviewed in detail by Sagae et al. [1], which reported clinical outcomes for small cohorts of patients with the majority having no measurable disease at initiation of chemotherapy (CT). Without randomized comparisons for defining platinum-refractory, resistant, or sensitive USC in the adjuvant setting, ascertaining therapeutic efficacy is indeed challenging. PFS is the current metric of choice for determining drug sensitivity for patients harboring tumors with reliable biomarkers or measurable disease. The use of such parameters for primary USC is limited and infrequently applicable, except in stage IV with macroscopic residual disease. The lag time requisite for microscopic refractory or resistant residual USC to become clinically detectable and afford reliable assessment of PFS is undefined. USC has an aggressive natural history, and the effectiveness of contemporary, adjuvant PbCT in the absence of measurable disease has not been established. Clinical trial data and current observations as well as the experiential transition in managing USC at our institution over the past 18 years were incentives to conduct a more detailed assessment of the effectiveness of adjuvant therapies for USC.
administered sequentially, although PbCT was infrequently “sandwiched” around EBRT. Pelvic EBRT was administered at 45 to 50 Gy in 5 to 7 weeks, with fields extended for paraaortic metastases. Of 190 patients who received adjuvant CT, treatment methods were not available for 8. PbCT, either cisplatin or carboplatin in standard doses, was given to the remaining 182 patients (94.8%) in combination with paclitaxel (n = 168) or doxorubicin (n = 7), or both (n = 7). Pathologic criteria used to diagnose USC typically included foci of papillary growth, high-grade cytologic findings, hob-nail cells, and cells exfoliated into luminal spaces. In cases where the diagnosis was suspected from high-grade cytologic findings alone, p53 over expression or loss, and strong, block-like p16 expression were used for confirmation. In specimens with mixed differentiation, the serous component was considered the principal histologic finding. Systematic histologic reassessment of specimens accrued before 2009 and intermittently thereafter was conducted by a senior gynecologic pathologist (G.L.K).
2. Methods
3. Results
The number of new cases of endometrial cancer is increasing in the United States [5]. In our tertiary care practice, the volume is also increasing, with the prevalence of USC (approximately 14%) somewhat higher than expected. From January 1, 1999, through June 30, 2016, 409 patients diagnosed with serous carcinoma or serous carcinoma with either endometrioid or clear cell carcinomas of documented uterine origin were eligible for inclusion in this study. Patients were excluded if they declined research authorization, had invasive synchronous cancers, or had received neoadjuvant CT. This study was approved by the Mayo Clinic Institutional Review Board. Only those patients who consented to use of their medical records for research were included in the analyses, in accordance with the Minnesota Statute for Use of Medical Information in Research. During the study period, surgical staging and cytoreduction followed by selective adjuvant therapies were the main therapies. The benchmarks used for staging were based on the 2009 International Federation of Gynecology and Obstetrics criteria for endometrial cancer [6]. During the first tertile of this study, the standardized treatment algorithm was adjusted periodically after outcomes were assessed. Beginning in 2004, specific surgical guidelines were developed for staging, and periodic quality assessments of the results were begun [7]. Shortly thereafter (first tertile of this study), we demonstrated a need for adjuvant vaginal brachytherapy (VBRT) in patients with early stage disease who had vaginal recurrences, grade 3 histologic findings, and lymphovascular space involvement [8]. The primary adjuvant therapy for stage I disease was VBRT ± PbCT; however, PbCT ± external beam radiotherapy (EBRT) was considered standard treatment for eligible patients with advanced-stage disease. Combined therapy was preferably
During the study period, 409 patients with USC met the inclusion criteria. Table 1 shows age at surgery, histologic findings, stage, and therapeutic interventions for early localized and advanced disease. Patients with stage III/IV disease more often had positive cytologic findings, tumor size N2 cm, myometrial invasion ≥50%, and adjuvant systemic therapy. Of the patients with stage III/IV disease, 64.9% underwent cytoreduction to no residual disease (R0) (87.4% had residual disease ≤1 cm [R1]). Of 218 patients with early stage disease, 209 had stage I disease. Postoperative adjuvant therapy for the patients with stage I disease included VBRT alone in 79 patients; PbCT alone, 8 patients; PbCT plus VBRT, 44 patients; and EBRT only, 4 patients. Fifty-six patients were observed only, and the initiation/completion of adjuvant therapy was not documented for 18 patients. Of the 209 patients with stage I disease, 26 (12.4%) had a documented progression at a median of 1.3 years after surgery (interquartile range [IQR], 0.8–2.9 years). The median duration of relevant clinical follow-up for the remaining 183 patients with stage I disease was 3.1 years (IQR, 0.7–7.7 years). Overall, the estimated 5-year PFS for patients with stage I disease was 81.3% (95% CI, 74.7–88.5). In multivariable analysis, adjuvant therapy was the strongest predictor of PFS; none of the other variables (Table 1) were significant after adjuvant therapy was included in the model (results not shown). Compared with observation only, clinical outcomes (PFS) were better after either VBRT only or PbCT ± VBRT (P = .004) (Fig. 1). The estimated 5-year PFS for observation only was 65.1% (95% CI, 51.3–82.7); VBRT only, 90.7% (95% CI, 83.1–99.0); and PbCT ± VBRT, 91.1% (95% CI, 81.3–100.0). Of the patients with advanced disease, 72 had stage IIIC disease. Of these patients, 67 (93.1%) had systemic lymphadenectomy, including
2.1. Statistical analyses Statistical analyses were performed using SAS version 9.3 software (SAS Institute Inc). PFS after surgery was estimated using the KaplanMeier method; the duration of follow-up for patients without a documented progression was censored at the date of their last relevant clinical follow-up. Comparisons of PFS between groups were evaluated using the log-rank test. Multivariable Cox proportional hazards regression models were fit to identify factors independently associated with progression. Progression rates (PRs), stratified by extent of residual disease and follow-up intervals, were calculated as the number of patients in each stratum, with progression divided by the total duration of follow-up during each follow-up interval and expressed per 1 personyear of follow-up. Exact 95% CIs were constructed, assuming the observed number of patients with progressions followed a Poisson distribution and the number of person-years was fixed. All calculated P values were 2-sided, and P values b .05 were considered statistically significant.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
3
Table 1 Patient characteristics. Characteristic
Overall (N = 409)
Early Stage (I/II) (n = 218)
Advanced Stage (III/IV) (n = 191)
Age at surgery, mean (SD), y Histologic findings, No. (%) Pure serous Mixed serous Unable to determine Cytologic findings, No. (%) Negative Positive Not sampled Tumor diameter, No. (%) ≤ 2 cm N 2 cm Unknown Myometrial invasion, No. (%) None b 50% ≥ 50% Extent of LND, No. (%) No LND Pelvic LND only Paraaortic LND only Pelvic and paraaortic LND FIGO stage (2009), No. (%) I II III IV Adjuvant therapy, No. (%) Observation only VBRT EBRT ± VBRT PbCT ± VBRT PbCT and EBRT ± VBRT Unknown Residual disease, No. (%) None (R0) ≤ 1 cm (R1) N 1 cm (R2)
69.3 (10.1)
69.5 (10.0)
69.0 (10.2)
269 (65.8) 117 (28.6) 23 (5.6)
135 (61.9) 70 (32.1) 13 (6.0)
134 (70.2) 47 (24.6) 10 (5.2)
239 (58.4) 134 (32.8) 36 (8.8)
175 (80.3) 21 (9.6) 22 (10.1)
64 (33.5) 113 (59.2) 14 (7.3)
90 (22.0) 303 (74.1) 16 (3.9)
61 (28.0) 146 (67.0) 11 (5.0)
29 (15.2) 157 (82.2) 5 (2.6)
113 (27.6) 171 (41.8) 125 (30.6)
80 (36.7) 103 (47.2) 35 (16.1)
33 (17.3) 68 (35.6) 90 (47.1)
83 (20.3) 56 (13.7) 4 (1.0) 266 (65.0)
42 (19.3) 39 (17.9) 1 (0.5) 136 (62.4)
41 (21.5) 17 (8.9) 3 (1.6) 130 (68.1)
209 (51.1) 9 (2.2) 85 (20.8) 106 (25.9)
209 (95.9) 9 (4.1) … …
… … 85 (44.5) 106 (55.5)
86 (21.0) 82 (20.0) 13 (3.2) 140 (34.2) 50 (12.2) 38 (9.3)
57 (26.1) 80 (36.7) 4 (1.8) 54 (24.8) … 23 (10.6)
29 (15.2) 2 (1.0) 9 (4.7) 86 (45.0) 50 (26.2) 15 (7.9)
342 (83.6) 43 (10.5) 24 (5.9)
218 (100.0) … …
124 (64.9) 43 (22.5) 24 (12.6)
Fig. 1. Progression-free survival of patients with FIGO stage I cancer (n = 209), stratified by adjuvant therapy: observation only, VBRT, and PbCT ± VBRT (85% of the PbCT cohort also received VBRT). EBRT only and unknown therapy are not shown. The 5-year PFS was 65.1% for patients (n = 56) who did not receive any adjuvant therapy (observation only), 90.7% for patients (n = 79) who received VBRT only, and 91.1% for patients (n = 52) who received PbCT ± VBRT. EBRT indicates external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; PbCT, platinum-based chemotherapy; PFS, progression-free survival; VBRT, vaginal brachytherapy.
Abbreviations: EBRT, external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; LND, lymphadenectomy; PbCT, platinum-based chemotherapy; R0, no residual disease; R1, residual disease ≤1 cm; R2, residual disease N1 cm; VBRT, vaginal brachytherapy.
in the pelvic and paraaortic basins, with dissection limited to the pelvis in 5 (6.9%) patients; 32 (44.4%) patients had stage IIIC1 disease and 40 (55.6%), stage IIIC2. CT alone or in combination with EBRT was primary adjuvant therapy for 58 (80.6%) patients. A documented progression occurred in 41 patients with stage IIIC disease at a median of 1.3 years after surgery (IQR, 0.7–2.0 years); the median duration of relevant clinical follow-up for the remaining 31 patients with stage IIIC disease was 3.0 years (IQR, 0.8–4.4 years). The estimated 5-year PFS for the patients with stage IIIC disease was 31.9% (95% CI, 21.3–48.0). Although the median PFS was different for patients in stages IIIC1 (36.5 months) and IIIC2 (18.4 months) (Fig. 2A), the PFS was nearly the same after 5 years (P = .20). Fifty-eight (80.6%) of the 72 patients with stage IIIC disease were treated with PbCT ± VBRT (n = 15) or PbCT + EBRT ± VBRT (n = 43); their overall 5-year PFS was 33.9%. A superior adjuvant treatment regimen was not discernible on univariate analysis (P = .13) (Fig. 2B). However, on multivariate analysis, both histologic findings and adjuvant therapy were independent predictors of PFS; specifically, patients with pure serous carcinoma had a significantly poorer prognosis than those with mixed serous carcinoma; patients observed only had a significantly poorer prognosis than those treated with PbCT + EBRT ± VBRT. Treatment failures for 38 of 41 stage IIIC patients with known sites of recurrence were hematogenous in 63.2% (11 sole site, 13 combined with other sites); lymphatic, 50.0% (9 sole site, 10 combined with other sites); peritoneal alone or in combination, 15.8%; and vagina alone or in combination, 10.5%.
Fig. 2. Progression-free survival of patients with FIGO stage IIIC cancer (n = 72). A, Stage IIIC1 and IIIC2. The 5-year PFS was 37.9% for IIIC1 (n = 32) and 27.7% for IIIC2 (n = 40). B, Adjuvant therapy. The median PFS was 7.3 months for no adjuvant therapy (observation only), 24.3 months for EBRT ± VBRT, 13.6 months for PbCT ± VBRT, and 36.5 months for CT and EBRT ± VBRT. EBRT indicates external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; PbCT, platinum-based chemotherapy; PFS, progression-free survival; VBRT, vaginal brachytherapy.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
4
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
Patients with stage IV disease (n = 106) composed 25.9% of the study population. At surgery, 38.7% of patients had complete cytoreduction (R0), and 40.6% had R1. Adjuvant therapy included PbCT ± EBRT for 72 patients; EBRT alone, 2 patients; VBRT alone, 1 patient; no documentation of initiation/completion of therapy, 10 patients; and opted for observation only, 21 patients. The primary reasons 24 patients did not elect cytotoxic therapy were extensive disease, frailty, or advanced age (mean, 75.3 years), or a combination. PbCT was included as an adjuvant therapy of choice for 72 patients (mean age, 67.2 years) who were candidates for CT. Specifically, PbCT included carboplatin (n = 63) or cisplatin (n = 5) in combination with taxane (n = 65) ± doxorubicin (n = 3) or doxorubicin (n = 3) ± gemcitabine (n = 1), with incomplete regimen documentation in 4 patients. Among those patients with stage IV disease who received PbCT ± EBRT (n = 72), 59 (81.9%) had a documented progression at a median of 0.7 years after surgery (IQR, 0.2–1.2 years). The median duration of relevant clinical follow-up for the remaining 13 patients in stage IV who received PbCT ± EBRT was 1.8 years (IQR, 0.6–4.5 years). The 5year PFS for patients with stage IV disease who received PbCT ± EBRT was 9.0% (95% CI, 3.8–21.5); the median PFS according to residual disease was 18.6 months for patients with R0; 8.0 months, R1; and 4.6 months, R2 (P = .008) (Fig. 3A). Most patients with R1 and R2 (residual disease N1 cm) tumors had a progression within 1 year of surgery, suggesting innate refractory/resistant disease. The PRs after surgery for R0 tumors at 1 to b2 years were similar to that of R1 tumors from 0 to b1 year (Fig. 3A). These observations suggest that determining whether an R0 tumor is sensitive to CT requires allowing for the time required for refractory microscopic disease to become clinically detectable. To study the potential parallels between R0 and R1 in patients with stage IV disease, interval PRs were determined as a function of personyears of follow-up, residual disease, and observation intervals (Table 2). The PR for 1 to b2 years of follow-up for R0 was 0.87; for R1 at 0 to b1 year, it was 1.27 (P = .31) (Table 2; Fig. 3B). These
observations, which allow for the 12-month lag-detection interval for R0 disease, suggest that most USCs in the R0 and R1 categories are innately refractory or resistant to PbCT. PR assessment using shorter clinical lag-detection intervals (6 and 9 months) for R0 disease suggested the 12-month interval from the date of surgery was statistically more representative, as was the elevated PR for R0 disease during the subsequent 12-month interval (1 to b2 years). Furthermore, the PRs were 0.13 per person-years' follow-up at ≥2 years in the R0 group and 0.14 for patients in the R1 group at or beyond 1 year (≥1 year) of follow-up (P = .95) (Table 2; Fig. 3C), suggesting that only a limited number of patients in these 2 cohorts may satisfy the criteria for sensitivity to PbCT. 4. Discussion In 2003, the senior author (K.C.P.) [9] penned an editorial for this journal entitled, “Uterine Papillary Serous Carcinoma: an Exigency for Clinical Trials,” referencing 2 featured articles that described retrospective outcomes assessments of USC managed at prominent institutions before the turn of the millennium [9–11]. During that period, 1987–2002, applicable treatment guidelines generally did not differentiate management according to histologic findings. Reported 5-year PFS rates were 65% to 70% for patients with stage I disease, 37% for stage III, and 0% for stage IV [10]. As chronicled in the review by Sagae et al. [1], multiple confirmatory studies expedited the recognition of the unique natural history of USC and the perceived requirements for comprehensive staging, cytoreduction, and aggressive adjuvant therapy. The management of USC at our institution likewise evolved with the implementation of quality assessment of surgical staging, cytoreduction, and liberalized use of PbCT. However, when outcomes of the above-referenced reports were compared with those observed in this study, we learned that the clinical outcomes have regrettably not improved substantially: 5-year PFS for patients in stage IIIC treated with PbCT was 33.9% and in stage IV, 9.0%. Perhaps the lack of substantial changes in PFS over these time
Fig. 3. Disease progression for patients with FIGO stage IV cancer who received PbCT (n = 72), according to no/microscopic residual disease (R0), ≤1 cm residual disease (R1), and N1 cm residual disease (R2). A, Progression-free survival according to residual disease. The 5-year PFS was 12.8% for R0, 8.2% for R1, and b10% for R2. B, Progression rate of R1 during the interval 0 to b1 year and of R0 during the interval of 1 to b2 years (bolded lines), showing similar rates of progression. C, Progression rates of R1 ≥ 1 year and R0 ≥ 2 years (bolded lines), showing similar rates of progression. FIGO indicates International Federation of Gynecology and Obstetrics.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx Table 2 PR of disease at surgery and follow-up interval for patients with FIGO Stage IV disease who received PbCT with or without EBRT. Follow-up, y
0 to b1 1 to b2 ≥2
PR per 1 person-year of follow-up (N = 72)a R0 (n = 27)
R1 (n = 35)
R2 (n = 10)
0.32 (7/21.91) (95% CI, 0.13–0.66) 0.87 (10/11.48) (95% CI, 0.42–1.60) 0.13 (3/22.70) (95% CI, 0.03–0.39)
1.27 (26/20.52) (95% CI, 0.83–1.86) 0.30 (2/6.71) (95% CI, 0.04–1.07) 0.09 (2/21.79) (95% CI, 0.01–0.33)
2.06 (9/4.36) (95% CI, 0.94–3.92) 0.00 (0/0.19) (95% CI, 0.00–19.05) …
Abbreviations: EBRT, external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; PbCT, platinum-based chemotherapy; PR, progression rate; R0, no/ microscopic residual disease; R1, residual disease ≤1 cm; R2, residual disease N1 cm. a The numbers in each cell denote the PR (number of patients with progression/total duration of follow-up during each follow-up interval).
intervals was predictable, considering that a survival benefit was not shown in the PbCT arms of 2 randomized clinical trials—1 for localized and 1 for advanced/recurrent USC [3,4]. In attempting to readdress the effectiveness of PbCT in early or advanced USC in the current study, we encountered the challenge of defining platinum sensitivity in a population in which 83.6% have no residual disease and infrequently harbor reliable tumor markers. Because the biology and natural history of USC differs from HGSOC, the standard definitions for refractory/resistant disease in USC might not have been applicable and were, thus, reexamined. We reasoned that establishing metrics that estimated the time interval for refractory or resistant microscopic disease to become clinically detectable would facilitate assessing platinum sensitivity in our population. In examining the PFS curve for stage IV R1, we found a dramatic attrition during the 0 to b1 year interval after surgery, with a notable flexion in the PFS curve after ≥1 year (0 to b1 year: 1.27 patients with progression/person-years; vs ≥1 year: 0.14 patients with progression/person-years). This finding strongly suggested that most of these patients' tumors were innately platinum refractory/resistant; only a limited number of patients whose tumors were potentially platinum sensitive were alive without disease 12 months after surgery. However, ambiguity exists for determining similar metrics for resistant and sensitive disease in patients with USC and microscopic residuum, which requires an undefined interval to become clinically detectable. The observed slope of the PFS curve for stage IV R0 during the 1 to b2 year interval after surgery appeared similar to the slope of the PFS curve for R1 during the initial year after surgery. The PR for stage IV R0 (0.87 patients with progression/person-years) during the 1 to b2 year interval after surgery did not differ significantly from R1 during the 0 to b1 year interval (P = .31). Furthermore, the PR after ≥2 years for R0 (0.13 patients with progression/person-years) was nearly identical to that of R1 after ≥1 year (0.14) (P = .95). On the basis of these observations, we believe that the lag interval for microscopic residuum to become clinically discernable is a requisite for assigning platinum refractory/resistant disease in USC. These observations suggest that USC treated with adjuvant PbCT should be designated as refractory/resistant disease if it progresses within 12 months after surgery in a patient with macroscopic residuum or within 24 months after surgery in a patient with microscopic residuum. Pragmatically assessing the effect of adjuvant therapy on clinical outcomes for stages I and III in this study was challenging. Although the impact of surgery was shown for stage I by the 5-year PFS (65.1%) in the observation-only cohort, curiously similar but significantly more favorable outcomes occurred for VBRT alone (5-year PFS, 90.7%) or PbCT ± VBRT (5-year PFS, 91.1%). The added value of PbCT to VBRT is questionable, considering that almost 85% of the stage I patients who received PbCT also received VBRT. Given the natural history of USC, the influence on outcomes of VBRT only is intriguing. During the latter part of the first tertile of this study, a detailed, periodic quality assessment algorithm for systematic lymphadenectomy was instituted, which may have
5
contributed to stage migration, thus resulting in a more genuine stage I cohort requiring minimal adjuvant therapy. The perceived added value of adjuvant PbCT in surgically confirmed stage I USC should ideally await results of definitive trials verifying its efficacy in advanced disease. In the 2003 report by Slomovitz et al. [10], adjuvant platinum, paclitaxel, or combinations were administered to 42% of patients with stage III cancer, with a 5-year PFS of 36.6%. In the present study, 81% of the 72 patients with stage IIIC cancer received PbCT and had a corresponding, disappointing 5-year PFS of 31.9%. Specifically, the 5-year PFS was 33.9% for the 58 patients in stage IIIC whose cancer was managed with PbCT ± EBRT ± VBRT. When the definition of innate PbCT refractory/resistant disease was adapted, as discussed above, an estimated 60% of stage IIIC2 cancers progressed within 2 years of surgery. In addition, 63% of stage IIIC treatment failures included a component of hematogenous dissemination. Despite a more aggressive approach with adjuvant PbCT therapy, frequently in combination with EBRT, outcomes did not differ from those of historical controls [10]. We believe it imperative to ask the following question when we consider our combined stage IIIC and IV USC observations: Is continuation of our current adjuvant treatment algorithms in the best interest of most patients with USC? Considering the significant PFS differential between pure and mixed USC histology in stage IIIC patients, stratification accordingly in future clinical USC trials would be prudent. The relative chemosensitivity of HGSOC and the relative chemoresistance of USC are exemplified in the disparity in mortality rates among patients with these malignant tumors and similar levels of disease dissemination [2,10]. The refractoriness of USC to contemporary PbCT appears to more closely simulate the 20% of HGSOC that likewise exhibit innate resistance to platinum therapy. These distinctions could portend potential similarities in the biology of these sitespecific, recalcitrant histologies. Hence, an exigency continues to exist for enhancing our understanding of the causal molecular mechanisms, for identifying predictive biomarkers of USC refractoriness to CT, and for identifying and incorporating novel therapeutic targets in phase I and II trials for patients with stage IIIC and IV USC. Assessing the effectiveness of PbCT in USC was the primary focus of this report. However, the study has limitations, including inherent biases from its retrospective nature and the inconsistent delivery of presumed optimal cumulative doses of CT. The latter limitation is typical of elderly patients (mean age of our cohort, 69 years), who require more frequent modifications and reductions in PbCT dose and administration. These concerns are minimized by the inclusion of consecutive cases from a large, single, institutional USC population. Other strengths include the standardization of surgical staging and cytoreduction with periodic quality assessment; uniform histologic criteria for USC diagnosis; routine administration of PbCT in eligible, consenting patients with advanced disease; and use of PFS or PRs to assess therapeutic efficacy.
Acknowledgments Role of the funding source No funding sources. Conflict of interest The authors have no conflicts of interest to declare. References [1] S. Sagae, N. Susumu, A.N. Viswanathan, D. Aoki, F.J. Backes, D.M. Provencher, et al., Gynecologic Cancer InterGroup (GCIG) consensus review for uterine serous carcinoma, Int. J. Gynecol. Cancer 24 (2014) S83–9. [2] D.S. Chi, E.L. Eisenhauer, J. Lang, J. Huh, L. Haddad, N.R. Abu-Rustum, et al., What is the optimal goal of primary cytoreductive surgery for bulky stage IIIC epithelial ovarian carcinoma (EOC)? Gynecol. Oncol. 103 (2006) 559–564.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
6
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
[3] M.E. Randall, V.L. Filiaci, H. Muss, N.M. Spirtos, R.S. Mannel, J. Fowler, et al., Randomized phase III trial of whole-abdominal irradiation versus doxorubicin and cisplatin chemotherapy in advanced endometrial carcinoma: a Gynecologic Oncology Group Study, J. Clin. Oncol. 24 (2006) 36–44. [4] T. Hogberg, M. Signorelli, C.F. de Oliveira, R. Fossati, A.A. Lissoni, B. Sorbe, et al., Sequential adjuvant chemotherapy and radiotherapy in endometrial cancer–results from two randomised studies, Eur. J. Cancer 46 (2010) 2422–2431. [5] C.A. Hamilton, D.S. Kapp, J.K. Chan, Clinical aspects of uterine papillary serous carcinoma, Curr. Opin. Obstet. Gynecol. 20 (2008) 26–33. [6] S. Pecorelli, Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium, Int. J. Gynaecol. Obstet. 105 (2009) 103–104. [7] J.N. Bakkum-Gamez, A. Mariani, S.C. Dowdy, A.L. Weaver, M.E. McGree, W.A. Cliby, et al., The impact of surgical guidelines and periodic quality assessment on the staging of endometrial cancer, Gynecol. Oncol. 123 (2011) 58–64.
[8] A. Mariani, S.C. Dowdy, G.L. Keeney, M.G. Haddock, T.G. Lesnick, K.C. Podratz, Predictors of vaginal relapse in stage I endometrial cancer, Gynecol. Oncol. 97 (2005) 820–827. [9] K.C. Podratz, A. Mariani, Uterine papillary serous carcinomas: the exigency for clinical trials, Gynecol. Oncol. 91 (2003) 461–462. [10] B.M. Slomovitz, T.W. Burke, P.J. Eifel, L.M. Ramondetta, E.G. Silva, A. Jhingran, et al., Uterine papillary serous carcinoma (UPSC): a single institution review of 129 cases, Gynecol. Oncol. 91 (2003) 463–469. [11] W.K. Huh, M. Powell, C.A. Leath 3rd, J.M. Straughn Jr., D.E. Cohn, M.A. Gold, et al., Uterine papillary serous carcinoma: comparisons of outcomes in surgical Stage I patients with and without adjuvant therapy, Gynecol. Oncol. 91 (2003) 470–475.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
Contents lists available at ScienceDirect
Gynecologic Oncology journal homepage: www.elsevier.com/locate/ygyno
Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy Lucia Tortorella a, Carrie L. Langstraat a,b, Amy L. Weaver c, Michaela E. McGree c, Jamie N. Bakkum-Gamez a, Sean C. Dowdy a, William A. Cliby a, Gary L. Keeney d, Mark E. Sherman e, Saravut J. Weroha b, Andrea Mariani a, Karl C. Podratz a,⁎ a
Division of Gynecologic Surgery, Mayo Clinic, Rochester, MN, USA Division of Medical Oncology, Mayo Clinic, Rochester, MN, USA c Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN, USA d Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, MN, USA e Department of Health Sciences Research, Mayo Clinic, Jacksonville, FL, USA b
H I G H L I G H T S • Most uterine serous carcinomas fail to respond to platinum-based chemotherapy. • Critical lag times precede declaration of platinum refractory microscopic disease. • Truncated survival/systemic failures compel a search for more effective therapy.
a r t i c l e
i n f o
Article history: Received 27 December 2017 Received in revised form 20 February 2018 Accepted 28 February 2018 Available online xxxx Keywords: Platinum-based chemotherapy Therapeutic efficacy Uterine serous carcinoma
a b s t r a c t Objective. Two randomized trials failed to demonstrate efficacy of platinum-based chemotherapy (PbCT) for uterine serous carcinoma (USC). Our objective was to reassess the value of PbCT for patients with microscopic residuum (R0). Methods. Progression-free survival (PFS) after surgery was analyzed for 409 patients and correlated with adjuvant therapies: vaginal brachytherapy (VBRT), external beam radiotherapy (EBRT), PbCT, or combinations. Results. The estimated 5-year PFS for stage I (n = 209) USC was 65.1% for observation only; 90.7%, VBRT only; and 91.1%, PbCT ± VBRT (85% received VBRT); VBRT significantly (P = .004) impacted PFS, but the added value of PbCT remains uncertain. Of 58 stage IIIC, PbCT-treated patients (±EBRT), 5-year PFS was 33.9%; most failures had a vascular disseminated component. Median PFS for 72 stage IV, PbCT-treated patients was 18.6 months for R0; 8.0, R1 ≤ 1 cm residual disease; and 4.6, R2 N 1 cm (P = .008). The progression rate (PR) during 1 to 2 year followup for R0 was similar to PR during 0–1 year follow-up for R1 (P = .31), suggesting recurrences in patients with R0 disease before 2 years are likely platinum resistant. PRs during follow-up were nearly identical for R0 ≥ 2 years and R1 ≥ 1 year (P = .95), presumably showing limited numbers of platinum-sensitive tumors. Conclusions. A comparison of PR for patients treated with PbCT for stage IV R0 and R1 disease suggested that a 1-year lag interval precedes clinical recognition of PbCT refractory/resistant R0 disease. Most patients treated with PbCT who had microscopic residuum had recurrences within 2 years (across stages), emphasizing the need for more effective therapy. © 2018 Published by Elsevier Inc.
Abbreviations: CT, chemotherapy; EBRT, external beam radiotherapy; EEC, endometrial endometrioid carcinoma; HGSOC, high-grade serous ovarian cancer; HR, hazard ratio; IQR, interquartile range; OS, overall survival; PbCT, platinum-based chemotherapy; PFS, progression-free survival; PR, progression rate; R0, no macroscopic residual disease; R1, residual disease ≤1 cm; R2, residual disease N1 cm; RT, radiotherapy; USC, uterine serous carcinoma; VBRT, vaginal brachytherapy. ⁎ Corresponding author at: Division of Gynecologic Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905, USA. E-mail address: [email protected] (K.C. Podratz).
1. Introduction The current therapeutic algorithm for advanced-stage, high-grade uterine epithelial cancers generally includes surgical staging, cytoreduction, and administration of adjuvant platinum-based chemotherapy (PbCT) [1]. These principles have been extrapolated from the management of advanced-stage, high-grade serous ovarian cancer (HGSOC), for which the response rate to PbCT is 80%, and median
https://doi.org/10.1016/j.ygyno.2018.02.022 0090-8258/© 2018 Published by Elsevier Inc.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
2
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
overall survival (OS) rates exceed 5 years in patients whose tumors are completely cytoreduced [2]. This experiential model has been advocated for both advanced endometrial endometrioid carcinoma (EEC) and uterine serous carcinoma (USC). However, in the protocol based on Gynecologic Oncology Group 122, a clinical trial that randomized patients with advanced and recurrent endometrial cancer to either abdominal irradiation or PbCT, a survival advantage was readily shown with PbCT for EEC (progression-free survival [PFS] hazard ratio [HR], 0.687; OS HR, 0.482) but not for USC (PFS HR, 0.909; OS HR, 1.025) [3]. Similarly, a prospective clinical trial of pelvic radiotherapy (RT) alone vs RT plus PbCT for patients with stage I/II endometrial cancer showed a survival advantage with added PbCT for patients with EEC but not for those with USC or clear cell carcinomas [4]. Nevertheless, the increasing use of PbCT for treating USC at our institution and others, even for stage I patients, presupposes an evolving standard of care with uncertain justification [1]. This transition in clinical practice patterns is based on numerous retrospective studies, reviewed in detail by Sagae et al. [1], which reported clinical outcomes for small cohorts of patients with the majority having no measurable disease at initiation of chemotherapy (CT). Without randomized comparisons for defining platinum-refractory, resistant, or sensitive USC in the adjuvant setting, ascertaining therapeutic efficacy is indeed challenging. PFS is the current metric of choice for determining drug sensitivity for patients harboring tumors with reliable biomarkers or measurable disease. The use of such parameters for primary USC is limited and infrequently applicable, except in stage IV with macroscopic residual disease. The lag time requisite for microscopic refractory or resistant residual USC to become clinically detectable and afford reliable assessment of PFS is undefined. USC has an aggressive natural history, and the effectiveness of contemporary, adjuvant PbCT in the absence of measurable disease has not been established. Clinical trial data and current observations as well as the experiential transition in managing USC at our institution over the past 18 years were incentives to conduct a more detailed assessment of the effectiveness of adjuvant therapies for USC.
administered sequentially, although PbCT was infrequently “sandwiched” around EBRT. Pelvic EBRT was administered at 45 to 50 Gy in 5 to 7 weeks, with fields extended for paraaortic metastases. Of 190 patients who received adjuvant CT, treatment methods were not available for 8. PbCT, either cisplatin or carboplatin in standard doses, was given to the remaining 182 patients (94.8%) in combination with paclitaxel (n = 168) or doxorubicin (n = 7), or both (n = 7). Pathologic criteria used to diagnose USC typically included foci of papillary growth, high-grade cytologic findings, hob-nail cells, and cells exfoliated into luminal spaces. In cases where the diagnosis was suspected from high-grade cytologic findings alone, p53 over expression or loss, and strong, block-like p16 expression were used for confirmation. In specimens with mixed differentiation, the serous component was considered the principal histologic finding. Systematic histologic reassessment of specimens accrued before 2009 and intermittently thereafter was conducted by a senior gynecologic pathologist (G.L.K).
2. Methods
3. Results
The number of new cases of endometrial cancer is increasing in the United States [5]. In our tertiary care practice, the volume is also increasing, with the prevalence of USC (approximately 14%) somewhat higher than expected. From January 1, 1999, through June 30, 2016, 409 patients diagnosed with serous carcinoma or serous carcinoma with either endometrioid or clear cell carcinomas of documented uterine origin were eligible for inclusion in this study. Patients were excluded if they declined research authorization, had invasive synchronous cancers, or had received neoadjuvant CT. This study was approved by the Mayo Clinic Institutional Review Board. Only those patients who consented to use of their medical records for research were included in the analyses, in accordance with the Minnesota Statute for Use of Medical Information in Research. During the study period, surgical staging and cytoreduction followed by selective adjuvant therapies were the main therapies. The benchmarks used for staging were based on the 2009 International Federation of Gynecology and Obstetrics criteria for endometrial cancer [6]. During the first tertile of this study, the standardized treatment algorithm was adjusted periodically after outcomes were assessed. Beginning in 2004, specific surgical guidelines were developed for staging, and periodic quality assessments of the results were begun [7]. Shortly thereafter (first tertile of this study), we demonstrated a need for adjuvant vaginal brachytherapy (VBRT) in patients with early stage disease who had vaginal recurrences, grade 3 histologic findings, and lymphovascular space involvement [8]. The primary adjuvant therapy for stage I disease was VBRT ± PbCT; however, PbCT ± external beam radiotherapy (EBRT) was considered standard treatment for eligible patients with advanced-stage disease. Combined therapy was preferably
During the study period, 409 patients with USC met the inclusion criteria. Table 1 shows age at surgery, histologic findings, stage, and therapeutic interventions for early localized and advanced disease. Patients with stage III/IV disease more often had positive cytologic findings, tumor size N2 cm, myometrial invasion ≥50%, and adjuvant systemic therapy. Of the patients with stage III/IV disease, 64.9% underwent cytoreduction to no residual disease (R0) (87.4% had residual disease ≤1 cm [R1]). Of 218 patients with early stage disease, 209 had stage I disease. Postoperative adjuvant therapy for the patients with stage I disease included VBRT alone in 79 patients; PbCT alone, 8 patients; PbCT plus VBRT, 44 patients; and EBRT only, 4 patients. Fifty-six patients were observed only, and the initiation/completion of adjuvant therapy was not documented for 18 patients. Of the 209 patients with stage I disease, 26 (12.4%) had a documented progression at a median of 1.3 years after surgery (interquartile range [IQR], 0.8–2.9 years). The median duration of relevant clinical follow-up for the remaining 183 patients with stage I disease was 3.1 years (IQR, 0.7–7.7 years). Overall, the estimated 5-year PFS for patients with stage I disease was 81.3% (95% CI, 74.7–88.5). In multivariable analysis, adjuvant therapy was the strongest predictor of PFS; none of the other variables (Table 1) were significant after adjuvant therapy was included in the model (results not shown). Compared with observation only, clinical outcomes (PFS) were better after either VBRT only or PbCT ± VBRT (P = .004) (Fig. 1). The estimated 5-year PFS for observation only was 65.1% (95% CI, 51.3–82.7); VBRT only, 90.7% (95% CI, 83.1–99.0); and PbCT ± VBRT, 91.1% (95% CI, 81.3–100.0). Of the patients with advanced disease, 72 had stage IIIC disease. Of these patients, 67 (93.1%) had systemic lymphadenectomy, including
2.1. Statistical analyses Statistical analyses were performed using SAS version 9.3 software (SAS Institute Inc). PFS after surgery was estimated using the KaplanMeier method; the duration of follow-up for patients without a documented progression was censored at the date of their last relevant clinical follow-up. Comparisons of PFS between groups were evaluated using the log-rank test. Multivariable Cox proportional hazards regression models were fit to identify factors independently associated with progression. Progression rates (PRs), stratified by extent of residual disease and follow-up intervals, were calculated as the number of patients in each stratum, with progression divided by the total duration of follow-up during each follow-up interval and expressed per 1 personyear of follow-up. Exact 95% CIs were constructed, assuming the observed number of patients with progressions followed a Poisson distribution and the number of person-years was fixed. All calculated P values were 2-sided, and P values b .05 were considered statistically significant.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
3
Table 1 Patient characteristics. Characteristic
Overall (N = 409)
Early Stage (I/II) (n = 218)
Advanced Stage (III/IV) (n = 191)
Age at surgery, mean (SD), y Histologic findings, No. (%) Pure serous Mixed serous Unable to determine Cytologic findings, No. (%) Negative Positive Not sampled Tumor diameter, No. (%) ≤ 2 cm N 2 cm Unknown Myometrial invasion, No. (%) None b 50% ≥ 50% Extent of LND, No. (%) No LND Pelvic LND only Paraaortic LND only Pelvic and paraaortic LND FIGO stage (2009), No. (%) I II III IV Adjuvant therapy, No. (%) Observation only VBRT EBRT ± VBRT PbCT ± VBRT PbCT and EBRT ± VBRT Unknown Residual disease, No. (%) None (R0) ≤ 1 cm (R1) N 1 cm (R2)
69.3 (10.1)
69.5 (10.0)
69.0 (10.2)
269 (65.8) 117 (28.6) 23 (5.6)
135 (61.9) 70 (32.1) 13 (6.0)
134 (70.2) 47 (24.6) 10 (5.2)
239 (58.4) 134 (32.8) 36 (8.8)
175 (80.3) 21 (9.6) 22 (10.1)
64 (33.5) 113 (59.2) 14 (7.3)
90 (22.0) 303 (74.1) 16 (3.9)
61 (28.0) 146 (67.0) 11 (5.0)
29 (15.2) 157 (82.2) 5 (2.6)
113 (27.6) 171 (41.8) 125 (30.6)
80 (36.7) 103 (47.2) 35 (16.1)
33 (17.3) 68 (35.6) 90 (47.1)
83 (20.3) 56 (13.7) 4 (1.0) 266 (65.0)
42 (19.3) 39 (17.9) 1 (0.5) 136 (62.4)
41 (21.5) 17 (8.9) 3 (1.6) 130 (68.1)
209 (51.1) 9 (2.2) 85 (20.8) 106 (25.9)
209 (95.9) 9 (4.1) … …
… … 85 (44.5) 106 (55.5)
86 (21.0) 82 (20.0) 13 (3.2) 140 (34.2) 50 (12.2) 38 (9.3)
57 (26.1) 80 (36.7) 4 (1.8) 54 (24.8) … 23 (10.6)
29 (15.2) 2 (1.0) 9 (4.7) 86 (45.0) 50 (26.2) 15 (7.9)
342 (83.6) 43 (10.5) 24 (5.9)
218 (100.0) … …
124 (64.9) 43 (22.5) 24 (12.6)
Fig. 1. Progression-free survival of patients with FIGO stage I cancer (n = 209), stratified by adjuvant therapy: observation only, VBRT, and PbCT ± VBRT (85% of the PbCT cohort also received VBRT). EBRT only and unknown therapy are not shown. The 5-year PFS was 65.1% for patients (n = 56) who did not receive any adjuvant therapy (observation only), 90.7% for patients (n = 79) who received VBRT only, and 91.1% for patients (n = 52) who received PbCT ± VBRT. EBRT indicates external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; PbCT, platinum-based chemotherapy; PFS, progression-free survival; VBRT, vaginal brachytherapy.
Abbreviations: EBRT, external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; LND, lymphadenectomy; PbCT, platinum-based chemotherapy; R0, no residual disease; R1, residual disease ≤1 cm; R2, residual disease N1 cm; VBRT, vaginal brachytherapy.
in the pelvic and paraaortic basins, with dissection limited to the pelvis in 5 (6.9%) patients; 32 (44.4%) patients had stage IIIC1 disease and 40 (55.6%), stage IIIC2. CT alone or in combination with EBRT was primary adjuvant therapy for 58 (80.6%) patients. A documented progression occurred in 41 patients with stage IIIC disease at a median of 1.3 years after surgery (IQR, 0.7–2.0 years); the median duration of relevant clinical follow-up for the remaining 31 patients with stage IIIC disease was 3.0 years (IQR, 0.8–4.4 years). The estimated 5-year PFS for the patients with stage IIIC disease was 31.9% (95% CI, 21.3–48.0). Although the median PFS was different for patients in stages IIIC1 (36.5 months) and IIIC2 (18.4 months) (Fig. 2A), the PFS was nearly the same after 5 years (P = .20). Fifty-eight (80.6%) of the 72 patients with stage IIIC disease were treated with PbCT ± VBRT (n = 15) or PbCT + EBRT ± VBRT (n = 43); their overall 5-year PFS was 33.9%. A superior adjuvant treatment regimen was not discernible on univariate analysis (P = .13) (Fig. 2B). However, on multivariate analysis, both histologic findings and adjuvant therapy were independent predictors of PFS; specifically, patients with pure serous carcinoma had a significantly poorer prognosis than those with mixed serous carcinoma; patients observed only had a significantly poorer prognosis than those treated with PbCT + EBRT ± VBRT. Treatment failures for 38 of 41 stage IIIC patients with known sites of recurrence were hematogenous in 63.2% (11 sole site, 13 combined with other sites); lymphatic, 50.0% (9 sole site, 10 combined with other sites); peritoneal alone or in combination, 15.8%; and vagina alone or in combination, 10.5%.
Fig. 2. Progression-free survival of patients with FIGO stage IIIC cancer (n = 72). A, Stage IIIC1 and IIIC2. The 5-year PFS was 37.9% for IIIC1 (n = 32) and 27.7% for IIIC2 (n = 40). B, Adjuvant therapy. The median PFS was 7.3 months for no adjuvant therapy (observation only), 24.3 months for EBRT ± VBRT, 13.6 months for PbCT ± VBRT, and 36.5 months for CT and EBRT ± VBRT. EBRT indicates external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; PbCT, platinum-based chemotherapy; PFS, progression-free survival; VBRT, vaginal brachytherapy.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
4
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
Patients with stage IV disease (n = 106) composed 25.9% of the study population. At surgery, 38.7% of patients had complete cytoreduction (R0), and 40.6% had R1. Adjuvant therapy included PbCT ± EBRT for 72 patients; EBRT alone, 2 patients; VBRT alone, 1 patient; no documentation of initiation/completion of therapy, 10 patients; and opted for observation only, 21 patients. The primary reasons 24 patients did not elect cytotoxic therapy were extensive disease, frailty, or advanced age (mean, 75.3 years), or a combination. PbCT was included as an adjuvant therapy of choice for 72 patients (mean age, 67.2 years) who were candidates for CT. Specifically, PbCT included carboplatin (n = 63) or cisplatin (n = 5) in combination with taxane (n = 65) ± doxorubicin (n = 3) or doxorubicin (n = 3) ± gemcitabine (n = 1), with incomplete regimen documentation in 4 patients. Among those patients with stage IV disease who received PbCT ± EBRT (n = 72), 59 (81.9%) had a documented progression at a median of 0.7 years after surgery (IQR, 0.2–1.2 years). The median duration of relevant clinical follow-up for the remaining 13 patients in stage IV who received PbCT ± EBRT was 1.8 years (IQR, 0.6–4.5 years). The 5year PFS for patients with stage IV disease who received PbCT ± EBRT was 9.0% (95% CI, 3.8–21.5); the median PFS according to residual disease was 18.6 months for patients with R0; 8.0 months, R1; and 4.6 months, R2 (P = .008) (Fig. 3A). Most patients with R1 and R2 (residual disease N1 cm) tumors had a progression within 1 year of surgery, suggesting innate refractory/resistant disease. The PRs after surgery for R0 tumors at 1 to b2 years were similar to that of R1 tumors from 0 to b1 year (Fig. 3A). These observations suggest that determining whether an R0 tumor is sensitive to CT requires allowing for the time required for refractory microscopic disease to become clinically detectable. To study the potential parallels between R0 and R1 in patients with stage IV disease, interval PRs were determined as a function of personyears of follow-up, residual disease, and observation intervals (Table 2). The PR for 1 to b2 years of follow-up for R0 was 0.87; for R1 at 0 to b1 year, it was 1.27 (P = .31) (Table 2; Fig. 3B). These
observations, which allow for the 12-month lag-detection interval for R0 disease, suggest that most USCs in the R0 and R1 categories are innately refractory or resistant to PbCT. PR assessment using shorter clinical lag-detection intervals (6 and 9 months) for R0 disease suggested the 12-month interval from the date of surgery was statistically more representative, as was the elevated PR for R0 disease during the subsequent 12-month interval (1 to b2 years). Furthermore, the PRs were 0.13 per person-years' follow-up at ≥2 years in the R0 group and 0.14 for patients in the R1 group at or beyond 1 year (≥1 year) of follow-up (P = .95) (Table 2; Fig. 3C), suggesting that only a limited number of patients in these 2 cohorts may satisfy the criteria for sensitivity to PbCT. 4. Discussion In 2003, the senior author (K.C.P.) [9] penned an editorial for this journal entitled, “Uterine Papillary Serous Carcinoma: an Exigency for Clinical Trials,” referencing 2 featured articles that described retrospective outcomes assessments of USC managed at prominent institutions before the turn of the millennium [9–11]. During that period, 1987–2002, applicable treatment guidelines generally did not differentiate management according to histologic findings. Reported 5-year PFS rates were 65% to 70% for patients with stage I disease, 37% for stage III, and 0% for stage IV [10]. As chronicled in the review by Sagae et al. [1], multiple confirmatory studies expedited the recognition of the unique natural history of USC and the perceived requirements for comprehensive staging, cytoreduction, and aggressive adjuvant therapy. The management of USC at our institution likewise evolved with the implementation of quality assessment of surgical staging, cytoreduction, and liberalized use of PbCT. However, when outcomes of the above-referenced reports were compared with those observed in this study, we learned that the clinical outcomes have regrettably not improved substantially: 5-year PFS for patients in stage IIIC treated with PbCT was 33.9% and in stage IV, 9.0%. Perhaps the lack of substantial changes in PFS over these time
Fig. 3. Disease progression for patients with FIGO stage IV cancer who received PbCT (n = 72), according to no/microscopic residual disease (R0), ≤1 cm residual disease (R1), and N1 cm residual disease (R2). A, Progression-free survival according to residual disease. The 5-year PFS was 12.8% for R0, 8.2% for R1, and b10% for R2. B, Progression rate of R1 during the interval 0 to b1 year and of R0 during the interval of 1 to b2 years (bolded lines), showing similar rates of progression. C, Progression rates of R1 ≥ 1 year and R0 ≥ 2 years (bolded lines), showing similar rates of progression. FIGO indicates International Federation of Gynecology and Obstetrics.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx Table 2 PR of disease at surgery and follow-up interval for patients with FIGO Stage IV disease who received PbCT with or without EBRT. Follow-up, y
0 to b1 1 to b2 ≥2
PR per 1 person-year of follow-up (N = 72)a R0 (n = 27)
R1 (n = 35)
R2 (n = 10)
0.32 (7/21.91) (95% CI, 0.13–0.66) 0.87 (10/11.48) (95% CI, 0.42–1.60) 0.13 (3/22.70) (95% CI, 0.03–0.39)
1.27 (26/20.52) (95% CI, 0.83–1.86) 0.30 (2/6.71) (95% CI, 0.04–1.07) 0.09 (2/21.79) (95% CI, 0.01–0.33)
2.06 (9/4.36) (95% CI, 0.94–3.92) 0.00 (0/0.19) (95% CI, 0.00–19.05) …
Abbreviations: EBRT, external beam radiotherapy; FIGO, International Federation of Gynecology and Obstetrics; PbCT, platinum-based chemotherapy; PR, progression rate; R0, no/ microscopic residual disease; R1, residual disease ≤1 cm; R2, residual disease N1 cm. a The numbers in each cell denote the PR (number of patients with progression/total duration of follow-up during each follow-up interval).
intervals was predictable, considering that a survival benefit was not shown in the PbCT arms of 2 randomized clinical trials—1 for localized and 1 for advanced/recurrent USC [3,4]. In attempting to readdress the effectiveness of PbCT in early or advanced USC in the current study, we encountered the challenge of defining platinum sensitivity in a population in which 83.6% have no residual disease and infrequently harbor reliable tumor markers. Because the biology and natural history of USC differs from HGSOC, the standard definitions for refractory/resistant disease in USC might not have been applicable and were, thus, reexamined. We reasoned that establishing metrics that estimated the time interval for refractory or resistant microscopic disease to become clinically detectable would facilitate assessing platinum sensitivity in our population. In examining the PFS curve for stage IV R1, we found a dramatic attrition during the 0 to b1 year interval after surgery, with a notable flexion in the PFS curve after ≥1 year (0 to b1 year: 1.27 patients with progression/person-years; vs ≥1 year: 0.14 patients with progression/person-years). This finding strongly suggested that most of these patients' tumors were innately platinum refractory/resistant; only a limited number of patients whose tumors were potentially platinum sensitive were alive without disease 12 months after surgery. However, ambiguity exists for determining similar metrics for resistant and sensitive disease in patients with USC and microscopic residuum, which requires an undefined interval to become clinically detectable. The observed slope of the PFS curve for stage IV R0 during the 1 to b2 year interval after surgery appeared similar to the slope of the PFS curve for R1 during the initial year after surgery. The PR for stage IV R0 (0.87 patients with progression/person-years) during the 1 to b2 year interval after surgery did not differ significantly from R1 during the 0 to b1 year interval (P = .31). Furthermore, the PR after ≥2 years for R0 (0.13 patients with progression/person-years) was nearly identical to that of R1 after ≥1 year (0.14) (P = .95). On the basis of these observations, we believe that the lag interval for microscopic residuum to become clinically discernable is a requisite for assigning platinum refractory/resistant disease in USC. These observations suggest that USC treated with adjuvant PbCT should be designated as refractory/resistant disease if it progresses within 12 months after surgery in a patient with macroscopic residuum or within 24 months after surgery in a patient with microscopic residuum. Pragmatically assessing the effect of adjuvant therapy on clinical outcomes for stages I and III in this study was challenging. Although the impact of surgery was shown for stage I by the 5-year PFS (65.1%) in the observation-only cohort, curiously similar but significantly more favorable outcomes occurred for VBRT alone (5-year PFS, 90.7%) or PbCT ± VBRT (5-year PFS, 91.1%). The added value of PbCT to VBRT is questionable, considering that almost 85% of the stage I patients who received PbCT also received VBRT. Given the natural history of USC, the influence on outcomes of VBRT only is intriguing. During the latter part of the first tertile of this study, a detailed, periodic quality assessment algorithm for systematic lymphadenectomy was instituted, which may have
5
contributed to stage migration, thus resulting in a more genuine stage I cohort requiring minimal adjuvant therapy. The perceived added value of adjuvant PbCT in surgically confirmed stage I USC should ideally await results of definitive trials verifying its efficacy in advanced disease. In the 2003 report by Slomovitz et al. [10], adjuvant platinum, paclitaxel, or combinations were administered to 42% of patients with stage III cancer, with a 5-year PFS of 36.6%. In the present study, 81% of the 72 patients with stage IIIC cancer received PbCT and had a corresponding, disappointing 5-year PFS of 31.9%. Specifically, the 5-year PFS was 33.9% for the 58 patients in stage IIIC whose cancer was managed with PbCT ± EBRT ± VBRT. When the definition of innate PbCT refractory/resistant disease was adapted, as discussed above, an estimated 60% of stage IIIC2 cancers progressed within 2 years of surgery. In addition, 63% of stage IIIC treatment failures included a component of hematogenous dissemination. Despite a more aggressive approach with adjuvant PbCT therapy, frequently in combination with EBRT, outcomes did not differ from those of historical controls [10]. We believe it imperative to ask the following question when we consider our combined stage IIIC and IV USC observations: Is continuation of our current adjuvant treatment algorithms in the best interest of most patients with USC? Considering the significant PFS differential between pure and mixed USC histology in stage IIIC patients, stratification accordingly in future clinical USC trials would be prudent. The relative chemosensitivity of HGSOC and the relative chemoresistance of USC are exemplified in the disparity in mortality rates among patients with these malignant tumors and similar levels of disease dissemination [2,10]. The refractoriness of USC to contemporary PbCT appears to more closely simulate the 20% of HGSOC that likewise exhibit innate resistance to platinum therapy. These distinctions could portend potential similarities in the biology of these sitespecific, recalcitrant histologies. Hence, an exigency continues to exist for enhancing our understanding of the causal molecular mechanisms, for identifying predictive biomarkers of USC refractoriness to CT, and for identifying and incorporating novel therapeutic targets in phase I and II trials for patients with stage IIIC and IV USC. Assessing the effectiveness of PbCT in USC was the primary focus of this report. However, the study has limitations, including inherent biases from its retrospective nature and the inconsistent delivery of presumed optimal cumulative doses of CT. The latter limitation is typical of elderly patients (mean age of our cohort, 69 years), who require more frequent modifications and reductions in PbCT dose and administration. These concerns are minimized by the inclusion of consecutive cases from a large, single, institutional USC population. Other strengths include the standardization of surgical staging and cytoreduction with periodic quality assessment; uniform histologic criteria for USC diagnosis; routine administration of PbCT in eligible, consenting patients with advanced disease; and use of PFS or PRs to assess therapeutic efficacy.
Acknowledgments Role of the funding source No funding sources. Conflict of interest The authors have no conflicts of interest to declare. References [1] S. Sagae, N. Susumu, A.N. Viswanathan, D. Aoki, F.J. Backes, D.M. Provencher, et al., Gynecologic Cancer InterGroup (GCIG) consensus review for uterine serous carcinoma, Int. J. Gynecol. Cancer 24 (2014) S83–9. [2] D.S. Chi, E.L. Eisenhauer, J. Lang, J. Huh, L. Haddad, N.R. Abu-Rustum, et al., What is the optimal goal of primary cytoreductive surgery for bulky stage IIIC epithelial ovarian carcinoma (EOC)? Gynecol. Oncol. 103 (2006) 559–564.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022
6
L. Tortorella et al. / Gynecologic Oncology xxx (2018) xxx–xxx
[3] M.E. Randall, V.L. Filiaci, H. Muss, N.M. Spirtos, R.S. Mannel, J. Fowler, et al., Randomized phase III trial of whole-abdominal irradiation versus doxorubicin and cisplatin chemotherapy in advanced endometrial carcinoma: a Gynecologic Oncology Group Study, J. Clin. Oncol. 24 (2006) 36–44. [4] T. Hogberg, M. Signorelli, C.F. de Oliveira, R. Fossati, A.A. Lissoni, B. Sorbe, et al., Sequential adjuvant chemotherapy and radiotherapy in endometrial cancer–results from two randomised studies, Eur. J. Cancer 46 (2010) 2422–2431. [5] C.A. Hamilton, D.S. Kapp, J.K. Chan, Clinical aspects of uterine papillary serous carcinoma, Curr. Opin. Obstet. Gynecol. 20 (2008) 26–33. [6] S. Pecorelli, Revised FIGO staging for carcinoma of the vulva, cervix, and endometrium, Int. J. Gynaecol. Obstet. 105 (2009) 103–104. [7] J.N. Bakkum-Gamez, A. Mariani, S.C. Dowdy, A.L. Weaver, M.E. McGree, W.A. Cliby, et al., The impact of surgical guidelines and periodic quality assessment on the staging of endometrial cancer, Gynecol. Oncol. 123 (2011) 58–64.
[8] A. Mariani, S.C. Dowdy, G.L. Keeney, M.G. Haddock, T.G. Lesnick, K.C. Podratz, Predictors of vaginal relapse in stage I endometrial cancer, Gynecol. Oncol. 97 (2005) 820–827. [9] K.C. Podratz, A. Mariani, Uterine papillary serous carcinomas: the exigency for clinical trials, Gynecol. Oncol. 91 (2003) 461–462. [10] B.M. Slomovitz, T.W. Burke, P.J. Eifel, L.M. Ramondetta, E.G. Silva, A. Jhingran, et al., Uterine papillary serous carcinoma (UPSC): a single institution review of 129 cases, Gynecol. Oncol. 91 (2003) 463–469. [11] W.K. Huh, M. Powell, C.A. Leath 3rd, J.M. Straughn Jr., D.E. Cohn, M.A. Gold, et al., Uterine papillary serous carcinoma: comparisons of outcomes in surgical Stage I patients with and without adjuvant therapy, Gynecol. Oncol. 91 (2003) 470–475.
Please cite this article as: L. Tortorella, et al., Uterine serous carcinoma: Reassessing effectiveness of platinum-based adjuvant therapy, Gynecol Oncol (2018), https://doi.org/10.1016/j.ygyno.2018.02.022