Med Oncol (2014) 31:88 DOI 10.1007/s12032-014-0088-3
ORIGINAL PAPER
Individual-level data on the relationships of progression-free survival and post-progression survival with overall survival in patients with advanced non-squamous non-small cell lung cancer patients who received second-line chemotherapy Hisao Imai • Keita Mori • Akira Ono • Hiroaki Akamatsu • Tetsuhiko Taira • Hirotsugu Kenmotsu • Tateaki Naito • Kyoichi Kaira • Haruyasu Murakami • Masahiro Endo • Takashi Nakajima • Toshiaki Takahashi Received: 26 February 2014 / Accepted: 16 June 2014 / Published online: 25 June 2014 Ó Springer Science+Business Media New York 2014
Abstract The effects of second-line chemotherapy on overall survival (OS) might be confounded by subsequent therapies in patients with non-small cell lung cancer (NSCLC). Given the lack of research in this area, we here examined whether progression-free survival (PFS) or postprogression survival (PPS) could serve as valid surrogate endpoints for OS after second-line chemotherapy in advanced NSCLC, using individual-level data. Between April 2009 and June 2011, 39 patients with advanced nonsquamous NSCLC who had received second-line chemotherapy following first-line chemotherapy treatment with cisplatin and pemetrexed were analysed. The relationships of PFS and PPS with OS were analysed at the individual level. Spearman rank correlation analyses and linear regression analyses showed that PPS was strongly associated with OS (r = 0.90, p \ 0.05, R2 = 0.85), whereas
H. Imai (&) A. Ono H. Akamatsu T. Taira H. Kenmotsu T. Naito K. Kaira H. Murakami T. Takahashi Division of Thoracic Oncology, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan e-mail: [email protected]; [email protected] K. Mori Clinical Trial Coordination Office, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan M. Endo Division of Diagnostic Radiology, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan T. Nakajima Division of Diagnostic Pathology, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan
PFS only moderately correlated with OS (r = 0.76, p \ 0.05, R2 = 0.50). Best response at third-line treatment and number of regimens employed after progression beyond second-line chemotherapy were significantly associated with PPS (p \ 0.05). Analysis of individuallevel data of patients treated with second-line chemotherapy suggested that PPS may be used as a surrogate for OS in patients with advanced non-squamous NSCLC with unknown oncogenic driver mutations and therefore limited options for subsequent chemotherapy. Moreover, our findings suggest that subsequent treatment after disease progression following second-line chemotherapy may greatly influence OS. However, these results should be validated in further large-scale studies. Keywords Non-small cell lung cancer Overall survival Post-progression survival Progression-free survival Second-line chemotherapy
Introduction Lung cancer is the most common cause of cancer-related mortality worldwide, with non-small cell lung cancer (NSCLC) accounting for approximately 85 % of all lung cancers [1]. Overall survival (OS) is considered the most reliable endpoint in cancer studies, and, accordingly, when studies can be conducted to adequately assess survival, it is usually the preferred endpoint [2]. This endpoint is precise, easy to measure, and can be documented by the date of death. Surrogate endpoints such as tumour response and progression-free survival (PFS) are other useful endpoints for phase II oncology clinical trials, partly because they can be measured earlier and more conveniently, and partly because they occur more frequently than the main
123
88
Page 2 of 7
endpoints of interest (e.g. OS), which are referred to as the true endpoints. In view of the growing number of drugs and drug combinations available for the treatment of NSCLC, the effects of first-line chemotherapy on OS might be confounded by these subsequent therapies [3]. Indeed, PFS improvements do not necessarily result in an improved OS in patients with NSCLC, as shown by recent randomized trials [4]. In recent years, similarly to for breast, ovarian, and colorectal cancers [5–7], a growing number of active compounds have become available for second- or third-line chemotherapy for advanced NSCLC. Although PFS following first-line chemotherapy has not yet been validated as a surrogate endpoint for OS, post-progression survival (PPS) has conversely been shown to be strongly associated with OS after first-, second-, and third-line chemotherapy for advanced NSCLC at the clinical trial level [8–10]. PPS has moreover been demonstrated to strongly correlate with OS specifically during the last decade (2002–2012), corresponding with the introduction of molecular targeted agents such as gefitinib and erlotinib as chemotherapeutic agents for advanced NSCLC [8, 9]. In 2009, a simple method for evaluating PPS was first reported, in which OS is expressed as the sum of PFS and PPS [2]. At the individual level, the effect of therapies administered after disease progression on survival is of high interest. We have recently shown that post-progression survival after first-line chemotherapy for advanced nonsquamous NSCLC highly correlates with OS at the individual level [11]. To date, the validation of surrogate measures for OS after second-line therapy at the individual level in patients with advanced NSCLC has not been reported. Therefore, examination of whether PFS or PPS could be valid surrogate endpoints for OS after second-line therapy in patients with advanced NSCLC using individual-level data might be of clinical importance. Platinum-based doublet chemotherapy is currently the standard treatment for advanced NSCLC, based on modest benefits in survival and quality of life as compared with best supportive care only [12–17]. Although many patients initially achieve clinical remission or disease control with first-line chemotherapy, most subsequently experience disease progression and eventually die of advanced NSCLC. We have previously examined second-line chemotherapy following cisplatin and pemetrexed combination chemotherapy, which is considered standard first-line chemotherapy for advanced NSCLC [17]. In patients who experience disease progression, either during or after firstline chemotherapy for advanced NSCLC, second-line chemotherapy has become established over the past decade as a standard option [18–21]. The reported median survival time (MST) of patients harbouring an epidermal growth factor receptor (EGFR) mutation treated with gefitinib,
123
Med Oncol (2014) 31:88
Fig. 1 Flow chart showing patient selection. ALK anaplastic lymphoma kinase, EGFR epidermal growth factor receptor, PFS progression-free survival
platinum, and pemetrexed or docetaxel is approximately 3 years, whereas the MST of patients without an EGFR mutation is only about 1 year [22]. For advanced NSCLC patients without oncogenic driver mutations, such as an EGFR mutation, the OS is generally shorter and options for subsequent chemotherapy are currently limited. Although many phase III trials have been reported for previously treated patients with advanced NSCLC, no reports regarding PPS are available for this group at an individual level. Thus, in the present study, we aimed to analyse the relationships of PFS and PPS with OS at the individual level in advanced non-squamous NSCLC patients having received second-line chemotherapy. The patients evaluated had unknown oncogenic driver mutations, and therefore, options for subsequent-line chemotherapy were limited. Moreover, we also explored the prognostic value of baseline and tumour characteristics for PPS.
Methods Patients Between April 2009 and June 2011, 82 patients with advanced non-squamous NSCLC treated with cisplatin and pemetrexed as first-line chemotherapy were enrolled in this study. The tumour responses were not evaluated in 4 patients, an unknown EGFR mutation status was noted in 2, an EGFR mutation was observed in 15, the anaplastic lymphoma kinase (ALK) fusion gene was identified in 5, and PFS data were censored in 6. Forty-one patients were treated with second-line chemotherapy, and PFS data were censored in 2 patients. In order to unify patient background, these 43 patients were excluded from the analyses. Moreover, patients receiving maintenance therapy were also not considered, resulting in data from 39 patients
Med Oncol (2014) 31:88
being included in the final analysis (Fig. 1). The study protocol was approved by the Institutional Review Board of Shizuoka Cancer Center and conducted in accordance with the Helsinki declaration of 1975 (as revised in 2008). All patients were treated with cisplatin (75 mg m-2 day-1) and pemetrexed (500 mg m-2 day-1 for 1 day) followed by 21 days without treatment as first-line chemotherapy. The second-line chemotherapy regimens were determined by the treating physicians. The best overall response and maximum tumour shrinkage were recorded as the tumour responses. Radiographic tumour responses were evaluated according to the Response Evaluation Criteria In Solid Tumours, version 1.1 as follows: complete response (CR), disappearance of all target lesions; partial response (PR), at least a 30 % decrease in the sum of the target lesion diameters with the summed baseline diameters as a reference; progressive disease (PD), at least a 20 % increase in the sum of the target lesion diameters with the smallest sum observed during the study serving as reference; and stable disease (SD), insufficient shrinkage to qualify as PR and insufficient expansion to qualify as PD [23]. PFS was calculated from the start of second-line treatment to the date of PD or death from any cause. OS was recorded from the first day of second-line treatment until death or was censored on the date of the last follow-up consultation. PPS was recorded as the time from tumour progression after second-line treatment until death or was censored on the date of the last follow-up consultation.
Page 3 of 7 88 Table 1 Baseline patient characteristics at the beginning of secondline treatment Characteristic Sex, male/female
29/10
Median age at treatment (years)
64 (47–76)
Performance status (PS) 0/1/C2
9/29/1
Histology Adenocarcinoma/others
36/3
Stage IIIB/IV
3/36
Number of second-line chemotherapy courses 1/2/3/4/5/6/C 7
10/9/3/8/0/3/6
Number of regimens after progression following second-line chemotherapy 0/1/2/3/4/5/6
15/12/5/4/1/1/1
Median (range)
1 (0–6)
Table 2 Chemotherapy regimens employed after progression following first-line chemotherapy Secondline Docetaxel
Thirdline
CFourthline
Total 27
18
9
0
Erlotinib (single agent)
0
3
4
7
Erlotinib ± investigational agent
2
2
0
4
Other single agent
Statistical analyses To examine whether PFS or PPS correlated with OS, we used Spearman rank correlation and linear regression analyses. To explore prognostic factors for PPS, the proportional hazards model with a stepwise regression procedure was applied, and hazard ratios (HR) and 95 % confidence intervals (CI) were estimated. Because the HR is defined for a 1-unit difference, some factors were converted to an appropriate scale unit. PPS values were compared using the log-rank test. A p B 0.05 was considered significant for all tests. The twotailed significance level was also set at 0.05. All statistical analyses were performed using JMP version 9.0 for Windows (SAS Institute, Cary, NC, USA).
Results
S-1
5
3
4
12
Amrubicin
0
5
5
10
3
1
9
13
Platinum combination
Others
11
1
1
13
Investigational agent
0
0
2
2
present study are shown in Table 1. Out of the 39 patients, 4, 14, and 17 patients showed PR, SD, and PD, respectively. The response rate was 10.2 %, and the disease control rate was 46.1 %. After progressing past second-line chemotherapy, 15 of the 39 patients did not receive subsequent chemotherapy, and the median number of follow-up therapeutic regimens was one (range 0–6 regimens). The chemotherapy regimens employed after the first-line chemotherapy regimen are shown in Table 2. The median PFS and OS were 2.4 and 8.4 months, respectively (Fig. 2a, b).
Patient characteristics and treatment efficacy Of the 39 patients included in the analyses, 35 patients died; the median follow-up time was 8.0 months (range 1.0–32.0 months). The characteristics of the 39 patients (median age, 64 years; range, 47–76 years) included in the
Relationship between overall survival and progressionfree and post-progression survival The relationships between OS and PFS and PPS are shown in Figs. 3a and b, respectively. PPS was strongly associated
123
88
Page 4 of 7
Med Oncol (2014) 31:88
Fig. 2 Kaplan–Meier plots showing a progression-free survival (PFS) and b overall survival (OS). Median progression-free survival: 2.4 months; Median overall survival: 8.4 months; median follow-up: 8.0 months
Fig. 3 Correlations between a overall survival and progression-free survival, and b between overall survival and post-progression survival. *The r values represent Spearman’s rank correlation coefficient. **The R2 values represent linear regression
with OS (r = 0.90, p \ 0.05, R2 = 0.85), based on Spearman’s rank correlation coefficient and linear regression, whereas PFS was moderately correlated with OS (r = 0.76, p \ 0.05, R2 = 0.50). Factors affecting post-progression survival Because PPS was found to be strongly associated with OS, the association between PPS and various clinical factors was subsequently assessed. In the univariate analysis (Table 3), the PS at the beginning and end of the secondline treatment, as well as the best response at the secondline treatment, best response following third-line treatment (non-PD/PD), and the number of regimens employed after progression beyond second-line chemotherapy were found to be associated with PPS (p \ 0.05). Next, multivariate analysis for PPS was conducted to clarify which clinical factors may affect PPS (Table 4), and we found that the best response at third-line treatment (non-PD/PD) and the
123
number of regimens employed after progression beyond second-line chemotherapy were significantly associated with PPS (p \ 0.05). The log-rank tests confirmed that PPS differed in patients according to their best response at third-line treatment (non-PD/PD) and the number of regimens employed after progression beyond second-line chemotherapy (log-rank tests, p \ 0.05 (Fig. 4a, b). According to the best response at third-line treatment (non-PD/PD), patients with non-PD had a median PPS of 13.5 months, as compared with 4.5 months for patients with PD (log-rank tests, p \ 0.001; Fig. 4a). Furthermore, according to the number of regimens employed after progression beyond second-line chemotherapy, the PPS for patients receiving no, 1, and C2 additional regimens was 2.4, 4.1, and 11.9 months, respectively (log-rank tests, p \ 0.001; Fig. 4b). These results remained consistent after adjustment in the multivariate Cox proportional hazard analysis (Table 4).
Med Oncol (2014) 31:88
Page 5 of 7 88
Table 3 Univariate Cox regression analysis of baseline patient characteristics Factors
Post-progression survival Hazard ratio
95 % CI
p value
Factors
Sex, male/female
1.23
0.59–2.80
0.58
Age (years) at the beginning of second-line treatment
1.00
0.96–1.04
0.98
PS at the beginning of second-line treatment
3.72
1.56–10.71
ORIGINAL PAPER
Individual-level data on the relationships of progression-free survival and post-progression survival with overall survival in patients with advanced non-squamous non-small cell lung cancer patients who received second-line chemotherapy Hisao Imai • Keita Mori • Akira Ono • Hiroaki Akamatsu • Tetsuhiko Taira • Hirotsugu Kenmotsu • Tateaki Naito • Kyoichi Kaira • Haruyasu Murakami • Masahiro Endo • Takashi Nakajima • Toshiaki Takahashi Received: 26 February 2014 / Accepted: 16 June 2014 / Published online: 25 June 2014 Ó Springer Science+Business Media New York 2014
Abstract The effects of second-line chemotherapy on overall survival (OS) might be confounded by subsequent therapies in patients with non-small cell lung cancer (NSCLC). Given the lack of research in this area, we here examined whether progression-free survival (PFS) or postprogression survival (PPS) could serve as valid surrogate endpoints for OS after second-line chemotherapy in advanced NSCLC, using individual-level data. Between April 2009 and June 2011, 39 patients with advanced nonsquamous NSCLC who had received second-line chemotherapy following first-line chemotherapy treatment with cisplatin and pemetrexed were analysed. The relationships of PFS and PPS with OS were analysed at the individual level. Spearman rank correlation analyses and linear regression analyses showed that PPS was strongly associated with OS (r = 0.90, p \ 0.05, R2 = 0.85), whereas
H. Imai (&) A. Ono H. Akamatsu T. Taira H. Kenmotsu T. Naito K. Kaira H. Murakami T. Takahashi Division of Thoracic Oncology, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan e-mail: [email protected]; [email protected] K. Mori Clinical Trial Coordination Office, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan M. Endo Division of Diagnostic Radiology, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan T. Nakajima Division of Diagnostic Pathology, Shizuoka Cancer Center, 1007 Shimonagakubo, Nagaizumi-chou, Suntou-gun, Shizuoka 411-8777, Japan
PFS only moderately correlated with OS (r = 0.76, p \ 0.05, R2 = 0.50). Best response at third-line treatment and number of regimens employed after progression beyond second-line chemotherapy were significantly associated with PPS (p \ 0.05). Analysis of individuallevel data of patients treated with second-line chemotherapy suggested that PPS may be used as a surrogate for OS in patients with advanced non-squamous NSCLC with unknown oncogenic driver mutations and therefore limited options for subsequent chemotherapy. Moreover, our findings suggest that subsequent treatment after disease progression following second-line chemotherapy may greatly influence OS. However, these results should be validated in further large-scale studies. Keywords Non-small cell lung cancer Overall survival Post-progression survival Progression-free survival Second-line chemotherapy
Introduction Lung cancer is the most common cause of cancer-related mortality worldwide, with non-small cell lung cancer (NSCLC) accounting for approximately 85 % of all lung cancers [1]. Overall survival (OS) is considered the most reliable endpoint in cancer studies, and, accordingly, when studies can be conducted to adequately assess survival, it is usually the preferred endpoint [2]. This endpoint is precise, easy to measure, and can be documented by the date of death. Surrogate endpoints such as tumour response and progression-free survival (PFS) are other useful endpoints for phase II oncology clinical trials, partly because they can be measured earlier and more conveniently, and partly because they occur more frequently than the main
123
88
Page 2 of 7
endpoints of interest (e.g. OS), which are referred to as the true endpoints. In view of the growing number of drugs and drug combinations available for the treatment of NSCLC, the effects of first-line chemotherapy on OS might be confounded by these subsequent therapies [3]. Indeed, PFS improvements do not necessarily result in an improved OS in patients with NSCLC, as shown by recent randomized trials [4]. In recent years, similarly to for breast, ovarian, and colorectal cancers [5–7], a growing number of active compounds have become available for second- or third-line chemotherapy for advanced NSCLC. Although PFS following first-line chemotherapy has not yet been validated as a surrogate endpoint for OS, post-progression survival (PPS) has conversely been shown to be strongly associated with OS after first-, second-, and third-line chemotherapy for advanced NSCLC at the clinical trial level [8–10]. PPS has moreover been demonstrated to strongly correlate with OS specifically during the last decade (2002–2012), corresponding with the introduction of molecular targeted agents such as gefitinib and erlotinib as chemotherapeutic agents for advanced NSCLC [8, 9]. In 2009, a simple method for evaluating PPS was first reported, in which OS is expressed as the sum of PFS and PPS [2]. At the individual level, the effect of therapies administered after disease progression on survival is of high interest. We have recently shown that post-progression survival after first-line chemotherapy for advanced nonsquamous NSCLC highly correlates with OS at the individual level [11]. To date, the validation of surrogate measures for OS after second-line therapy at the individual level in patients with advanced NSCLC has not been reported. Therefore, examination of whether PFS or PPS could be valid surrogate endpoints for OS after second-line therapy in patients with advanced NSCLC using individual-level data might be of clinical importance. Platinum-based doublet chemotherapy is currently the standard treatment for advanced NSCLC, based on modest benefits in survival and quality of life as compared with best supportive care only [12–17]. Although many patients initially achieve clinical remission or disease control with first-line chemotherapy, most subsequently experience disease progression and eventually die of advanced NSCLC. We have previously examined second-line chemotherapy following cisplatin and pemetrexed combination chemotherapy, which is considered standard first-line chemotherapy for advanced NSCLC [17]. In patients who experience disease progression, either during or after firstline chemotherapy for advanced NSCLC, second-line chemotherapy has become established over the past decade as a standard option [18–21]. The reported median survival time (MST) of patients harbouring an epidermal growth factor receptor (EGFR) mutation treated with gefitinib,
123
Med Oncol (2014) 31:88
Fig. 1 Flow chart showing patient selection. ALK anaplastic lymphoma kinase, EGFR epidermal growth factor receptor, PFS progression-free survival
platinum, and pemetrexed or docetaxel is approximately 3 years, whereas the MST of patients without an EGFR mutation is only about 1 year [22]. For advanced NSCLC patients without oncogenic driver mutations, such as an EGFR mutation, the OS is generally shorter and options for subsequent chemotherapy are currently limited. Although many phase III trials have been reported for previously treated patients with advanced NSCLC, no reports regarding PPS are available for this group at an individual level. Thus, in the present study, we aimed to analyse the relationships of PFS and PPS with OS at the individual level in advanced non-squamous NSCLC patients having received second-line chemotherapy. The patients evaluated had unknown oncogenic driver mutations, and therefore, options for subsequent-line chemotherapy were limited. Moreover, we also explored the prognostic value of baseline and tumour characteristics for PPS.
Methods Patients Between April 2009 and June 2011, 82 patients with advanced non-squamous NSCLC treated with cisplatin and pemetrexed as first-line chemotherapy were enrolled in this study. The tumour responses were not evaluated in 4 patients, an unknown EGFR mutation status was noted in 2, an EGFR mutation was observed in 15, the anaplastic lymphoma kinase (ALK) fusion gene was identified in 5, and PFS data were censored in 6. Forty-one patients were treated with second-line chemotherapy, and PFS data were censored in 2 patients. In order to unify patient background, these 43 patients were excluded from the analyses. Moreover, patients receiving maintenance therapy were also not considered, resulting in data from 39 patients
Med Oncol (2014) 31:88
being included in the final analysis (Fig. 1). The study protocol was approved by the Institutional Review Board of Shizuoka Cancer Center and conducted in accordance with the Helsinki declaration of 1975 (as revised in 2008). All patients were treated with cisplatin (75 mg m-2 day-1) and pemetrexed (500 mg m-2 day-1 for 1 day) followed by 21 days without treatment as first-line chemotherapy. The second-line chemotherapy regimens were determined by the treating physicians. The best overall response and maximum tumour shrinkage were recorded as the tumour responses. Radiographic tumour responses were evaluated according to the Response Evaluation Criteria In Solid Tumours, version 1.1 as follows: complete response (CR), disappearance of all target lesions; partial response (PR), at least a 30 % decrease in the sum of the target lesion diameters with the summed baseline diameters as a reference; progressive disease (PD), at least a 20 % increase in the sum of the target lesion diameters with the smallest sum observed during the study serving as reference; and stable disease (SD), insufficient shrinkage to qualify as PR and insufficient expansion to qualify as PD [23]. PFS was calculated from the start of second-line treatment to the date of PD or death from any cause. OS was recorded from the first day of second-line treatment until death or was censored on the date of the last follow-up consultation. PPS was recorded as the time from tumour progression after second-line treatment until death or was censored on the date of the last follow-up consultation.
Page 3 of 7 88 Table 1 Baseline patient characteristics at the beginning of secondline treatment Characteristic Sex, male/female
29/10
Median age at treatment (years)
64 (47–76)
Performance status (PS) 0/1/C2
9/29/1
Histology Adenocarcinoma/others
36/3
Stage IIIB/IV
3/36
Number of second-line chemotherapy courses 1/2/3/4/5/6/C 7
10/9/3/8/0/3/6
Number of regimens after progression following second-line chemotherapy 0/1/2/3/4/5/6
15/12/5/4/1/1/1
Median (range)
1 (0–6)
Table 2 Chemotherapy regimens employed after progression following first-line chemotherapy Secondline Docetaxel
Thirdline
CFourthline
Total 27
18
9
0
Erlotinib (single agent)
0
3
4
7
Erlotinib ± investigational agent
2
2
0
4
Other single agent
Statistical analyses To examine whether PFS or PPS correlated with OS, we used Spearman rank correlation and linear regression analyses. To explore prognostic factors for PPS, the proportional hazards model with a stepwise regression procedure was applied, and hazard ratios (HR) and 95 % confidence intervals (CI) were estimated. Because the HR is defined for a 1-unit difference, some factors were converted to an appropriate scale unit. PPS values were compared using the log-rank test. A p B 0.05 was considered significant for all tests. The twotailed significance level was also set at 0.05. All statistical analyses were performed using JMP version 9.0 for Windows (SAS Institute, Cary, NC, USA).
Results
S-1
5
3
4
12
Amrubicin
0
5
5
10
3
1
9
13
Platinum combination
Others
11
1
1
13
Investigational agent
0
0
2
2
present study are shown in Table 1. Out of the 39 patients, 4, 14, and 17 patients showed PR, SD, and PD, respectively. The response rate was 10.2 %, and the disease control rate was 46.1 %. After progressing past second-line chemotherapy, 15 of the 39 patients did not receive subsequent chemotherapy, and the median number of follow-up therapeutic regimens was one (range 0–6 regimens). The chemotherapy regimens employed after the first-line chemotherapy regimen are shown in Table 2. The median PFS and OS were 2.4 and 8.4 months, respectively (Fig. 2a, b).
Patient characteristics and treatment efficacy Of the 39 patients included in the analyses, 35 patients died; the median follow-up time was 8.0 months (range 1.0–32.0 months). The characteristics of the 39 patients (median age, 64 years; range, 47–76 years) included in the
Relationship between overall survival and progressionfree and post-progression survival The relationships between OS and PFS and PPS are shown in Figs. 3a and b, respectively. PPS was strongly associated
123
88
Page 4 of 7
Med Oncol (2014) 31:88
Fig. 2 Kaplan–Meier plots showing a progression-free survival (PFS) and b overall survival (OS). Median progression-free survival: 2.4 months; Median overall survival: 8.4 months; median follow-up: 8.0 months
Fig. 3 Correlations between a overall survival and progression-free survival, and b between overall survival and post-progression survival. *The r values represent Spearman’s rank correlation coefficient. **The R2 values represent linear regression
with OS (r = 0.90, p \ 0.05, R2 = 0.85), based on Spearman’s rank correlation coefficient and linear regression, whereas PFS was moderately correlated with OS (r = 0.76, p \ 0.05, R2 = 0.50). Factors affecting post-progression survival Because PPS was found to be strongly associated with OS, the association between PPS and various clinical factors was subsequently assessed. In the univariate analysis (Table 3), the PS at the beginning and end of the secondline treatment, as well as the best response at the secondline treatment, best response following third-line treatment (non-PD/PD), and the number of regimens employed after progression beyond second-line chemotherapy were found to be associated with PPS (p \ 0.05). Next, multivariate analysis for PPS was conducted to clarify which clinical factors may affect PPS (Table 4), and we found that the best response at third-line treatment (non-PD/PD) and the
123
number of regimens employed after progression beyond second-line chemotherapy were significantly associated with PPS (p \ 0.05). The log-rank tests confirmed that PPS differed in patients according to their best response at third-line treatment (non-PD/PD) and the number of regimens employed after progression beyond second-line chemotherapy (log-rank tests, p \ 0.05 (Fig. 4a, b). According to the best response at third-line treatment (non-PD/PD), patients with non-PD had a median PPS of 13.5 months, as compared with 4.5 months for patients with PD (log-rank tests, p \ 0.001; Fig. 4a). Furthermore, according to the number of regimens employed after progression beyond second-line chemotherapy, the PPS for patients receiving no, 1, and C2 additional regimens was 2.4, 4.1, and 11.9 months, respectively (log-rank tests, p \ 0.001; Fig. 4b). These results remained consistent after adjustment in the multivariate Cox proportional hazard analysis (Table 4).
Med Oncol (2014) 31:88
Page 5 of 7 88
Table 3 Univariate Cox regression analysis of baseline patient characteristics Factors
Post-progression survival Hazard ratio
95 % CI
p value
Factors
Sex, male/female
1.23
0.59–2.80
0.58
Age (years) at the beginning of second-line treatment
1.00
0.96–1.04
0.98
PS at the beginning of second-line treatment
3.72
1.56–10.71
