Factors favouring long‑term survival following recurrence in ovarian cancer

Introduction

The median post-progression survival (PPS) for ovarian cancer patients is ~2 years (1). Ovarian cancer is a well-known chemosensitive malignancy. However, once the cancer relapses, it may be difficult to treat (1). Although a proportion of the patients achieve long-term PPS, the determining clinicopathological factors have yet to be fully elucidated.

It is considered that longer progression-free survival (PFS) may prolong overall survival (OS) based on the theory of reducing the number of cells available for subsequent mutation (2). Another study reported that the time-to-relapse is the most important prognostic factor in ovarian cancer, as subsequent chemotherapy regimens and the response to subsequent chemotherapy are determined based on this time interval (3). It was previously reported that secondary debulking surgery (SDS) was an effective treatment for recurrent ovarian carcinoma; however, the criteria for selecting SDS candidates remain unclear (4).

In the present study, the clinicopathological characteristics of patients achieving a longer PPS were analyzed in comparison with those of patients who succumbed to the disease earlier.

Patients and methods

Patient selection

Among patients with ovarian cancer who underwent first-line therapy between 1995 and 2006, those who developed relapse until the end of 2012 at the National Defense Medical College Hospital (Tokorozawa, Japan) were enrolled in the present study. The patients who achieved a 5-year PPS after the first recurrence or progression were defined as long-term survivors (LS), whereas those who succumbed to relapsed ovarian cancer within 5 years after the first recurrence or progression were defined as short-term survivors (SS). The patients who succumbed to other diseases and those with insufficient follow-up data were excluded (Fig. 1). The aim of this study was to identify differences in the clinicopathological characteristics between LS and SS based on the data acquired until the end of 2012.

Figure 1. Flow chart of the patient selection process.

Survival

In this study, OS was defined as the time from initial diagnosis of the disease to death, and PFS was defined as the time from initial diagnosis of the disease to diagnosis of clinical relapse, progression, or death. PPS was measured from the time of the first recurrence or progression to death. Platinum-free interval (PFI) was defined as the time between the end of the platinum-based chemotherapy and the first occurrence of relapse.

Treatment and follow-up

The patients were classified based on the International Federation of Gynecology and Obstetrics (FIGO) 1988 stage classification. Primary therapy included primary debulking surgery + adjuvant chemotherapy, neoadjuvant chemotherapy + interval debulking surgery, and first-line chemotherapy alone for inoperable patients. All the patients received platinum agents as first-line or adjuvant chemotherapy. SDS was only considered for patients whose identified recurrent tumors were expected to be completely resectable based on the clinician's discretion. For follow-up, physical and sonography examinations with carbohydrate antigen 125 (CA125) levels were evaluated every 3 months for 2 years. After 2 years, these examinations were performed every 3–6 months for at least 5 years. Computed tomography images were routinely checked every 6–12 months, or when the examinations raised the suspicion of recurrence. The Response Evaluation Criteria in Solid Tumors (RECIST), version 1.1 (5) was used to evaluate the effectiveness of the treatment. The Gynecologic Cancer Intergroup CA125 criteria were not used for diagnosis of recurrence (6). Performance status (PS) was evaluated based on the Eastern Cooperative Oncology Group (ECOG) PS system.

The clinical variables evaluated were as follows: Age at primary therapy, PS at primary therapy, FIGO stage, histology, residual tumor (RT) after primary therapy, PFI, age at recurrence, PS at recurrence, recurrent site, SDS, second-line chemotherapy regimens, response to second-line chemotherapy and number of chemotherapy regimens after the first recurrence.

Statistical analysis

Statistical analysis was performed using the JMP 10.0.0 software package (SAS Institute, Inc., Tokyo, Japan). The χ2 test and Mann-Whitney U test were used to evaluate differences in patient characteristics. Multivariable analyses were based on the logistic regression method and Kaplan-Meier survival curves of OS, PFS and PPS were compared using the log-rank test. Statistical significance was defined as P<0.05.

This study was approved by the Institutional Review Board of the National Defense Medical College.

Results

Patient characteristics

Among a total of 120 patients with ovarian cancer, 86 developed relapse, of whom 21 patients were excluded from the present study (2 patients succumbed to another disease and data were insufficient for 19 patients). Finally, 65 patients met the inclusion criteria and were enrolled in the study: 18 patients (28%) were classified as LS and 47 patients (72%) as SS (Fig. 1).

Patient survival

Kaplan-Meier survival curves for OS, PFS and PPS are presented in Fig. 2. The median PPS in LS and SS was 82.5 months (range, 60–122 months) and 15 months (range, 3–59 months), respectively (P<0.01, Fig. 2A). The median OS was 102.5 months in LS (range, 71–162 months) and 31 months in SS (range, 4–123 months) (P<0.01, Fig. 2B). In addition, the median PFS after primary therapy in LS and SS was 15.5 months (range, 71–162 months) and 4 months (range, 1–91 months), respectively (P=0.055, Fig. 2C).

Figure 2. Kaplan-Meier survival curves of the patients. (A) Post-progression survival (PPS) curves of long-term survivors (LS) and short-term survivors (SS). (B) Overall survival (OS) curves of LS and SS. (C) Progression-free survival (PFS) curves of LS and SS.

Patient characteristics

The clinicopathological characteristics of the patients are summarized in Table I. Significant differences between the two groups included RT after primary therapy (P<0.01), PFI (P=0.01), PS at recurrence (P<0.01), SDS (P=0.03) and favourable response to second-line chemotherapy (P<0.01).

Table I. Characteristics of the patients (n=65).

Table I.

Characteristics of the patients (n=65).

	Long-term survivors, n (%)	Short-term survivors, n (%)
Characteristics	(n=18)	(n=47)	P-value
Age at primary therapy (years)
≤54	12 (67)	20 (43)	0.08
>54	6 (33)	27 (57)
PS at primary therapy
0	12 (67)	30 (64)	0.83
≥1	6 (33)	17 (36)
FIGO stage
I	5 (28)	7 (15)	0.23
II–IV	13 (72)	40 (85)
Histology
Serous	6 (33)	18 (38)	0.14
Others	12 (67)	29 (62)
RT after primary therapy
Yes	1 (6)	20 (43)	<0.01
No	17 (94)	27 (57)
PFI at first recurrence (months)
<6	3 (17)	24 (51)	0.01
≥6	15 (83)	23 (49)
Age at first recurrence (years)
≤57	11 (61)	21 (45)	0.23
>57	7 (39)	26 (55)
PS at first recurrence
0	17 (94)	27 (57)	<0.01
≥1	1 (6)	20 (43)
Recurrent site
Distant metastasis	4 (22)	7 (15)	0.48
Others	14 (78)	40 (85)
Secondary debulking surgery
Yes	6 (33)	5 (11)	0.03
No	12 (67)	42 (89)
Second-line chemotherapy
Platinum-based	16 (89)	35 (74)	0.18
Others	2 (11)	12 (26)
Response to second-line chemotherapy
CR/PR	8 (44)	5 (11)	<0.01
SD/PD	10 (56)	42 (89)
Number of regimens after the first recurrence
≤2	8 (44)	31 (66)	0.11
≥3	10 (56)	16 (34)

[i] PS, performance status; FIGO, International Federation of Gynecology and Obstetrics; RT, residual tumor; PFI, platinum-free interval; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.

By multivariate analysis, SDS (OR=13.3; 95% CI: 1.39–226.7), favourable response to second-line chemotherapy (OR=16.5; 95% CI: 1.84–313.4) and receiving ≥3 regimens after the first recurrence (OR=9.01; 95% CI: 1.28–117.7) were identified as factors associated with long-term PPS (Table II).

Table II. Multivariate analysis for predictive factors of long-term survivors.

Table II.

Multivariate analysis for predictive factors of long-term survivors.

Variables	OR	95% CI	P-value
Age at primary therapy (years)
≤54	2.44	0.18–45.5	0.50
>54	1
PS at primary therapy
0	1.30	0.19–11.0	0.79
≥1	1
FIGO stage
I	9.69	0.87–168.6	0.07
II–IV	1
Histology
Serous	8.11	0.99–105.8	0.06
Others	1
RT after primary therapy
Yes	0.53	0.02–6.98	0.64
No	1
PFI at first recurrence (months)
<6	0.32	0.03–2.45	0.33
≥6	1
Age at first recurrence (years)
≤57	1.05	0.05–17.6	0.97
>57	1
PS at first recurrence
0	2.36	0.23–58.6	0.49
≥1	1
Recurrent site
Distant metastasis	2.85	0.31–33.7	0.35
Others	1
Secondary debulking surgery
Yes	13.3	1.39–226.7	0.03
No	1
Second-line chemotherapy
Platinum-based	3.68	0.40–57.2	0.26
Others	1
Response to second-line chemotherapy
CR/PR	16.5	1.84–313.4	0.01
SD/PD	1
Number of regimens after the first recurrence
≥3	9.01	1.28–117.7	0.03
≤2	1

[i] PS, performance states; FIGO, International Federation of Gynecology and Obstetrics; RT, residual tumor; PFI, platinum-free interval; OR, odds ratio; CI, confidence interval; CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease.

Discussion

Recent studies reported that PFS is not associated with OS, particularly in patients with advanced ovarian cancer (2,7,8). Shimokawa et al suggested that PPS rather than PFS was more significantly associated with OS (7). Additionally, another study reported that percentage gains in PFS are not associated with percentage gains in PPS (2). Our study also suggested that PFS was not associated with PPS. For this reason, although primary therapy for ovarian cancer is important, appropriate treatment after relapse may be more important for improving OS, in addition to prolonging PPS of ovarian cancer patients.

In the present study, SDS was significantly associated with LS after recurrence. The effectiveness of SDS has been long discussed, not only for platinum-sensitive, but also platinum-resistant recurrence (9–12). Some studies achieved long median OS (>50 months) after recurrence (11,12). However, the criteria for SDS candidacy have not fully established; therefore, the benefit of surgery for patients with relapsed ovarian cancer may be limited (13). Recently, selection criteria for operable patients with recurrent ovarian cancer were suggested by the DESKTOP trial: i) Good PS, ii) no or small volume of ascites at recurrence, and iii) no gross residual disease after primary surgery (14). Another study suggested that the predictive factors for complete resection at SDS included FIGO stage, complete primary surgery, PFI, PS, CA125 value at recurrence and ascites at recurrence (15). In the present study, none of the patients who underwent SDS had ascites, but some patients who had not met these criteria still achieved long-term survival. It is suggested that the selection of candidates for SDS requires further discussion.

The response to chemotherapy after recurrence also affected OS after recurrence. The response rates to second-line chemotherapy are generally different according to the platinum-sensitivity status: 20–25% for platinum-sensitive cases, and 10–20% for platinum-resistant cases (16,17). Our results suggested that the patients who exhibited complete or partial response to second-line regimens may become LS. Additionally, multiple lines of chemotherapy after recurrence also affected the PPS. Previous studies suggested the effectiveness of multiple lines of chemotherapy in ovarian cancer (18,19). This fact may be associated with homologous recombinant deficiency (HRD), as HRD has been suggested to be associated with chemosensitivity and longer OS (20). Generally, the duration of response to second-line regimens is shorter compared with that of primary chemotherapy (1). However, patients with recurrent disease may achieve longer OS by receiving several lines of chemotherapy. Hoskins et al reported that some selected patients based on PS and interval of chemotherapy may benefit from successive chemotherapy (18). Further studies are required to confirm the effectiveness of multiple lines of chemotherapy.

In conclusion, three factors favouring long PPS in ovarian cancer were identified, namely SDS, favourable response to second-line chemotherapy and ≥3 chemotherapy regimens after the first recurrence. Appropriate post-recurrence treatment is crucial for longer PPS. Although further analyses are required to evaluate the clinical significance of these factors, our study revealed potential clinicopathological markers favouring long-term PPS in recurrent ovarian cancer.

References