Efficacy and safety comparison of nabpaclitaxel plus S-1 and gemcitabine plus S-1 as first-line chemotherapy for metastatic pancreatic cancer

Efficacy and safety comparison of nabpaclitaxel plus S-1 and gemcitabine plus S-1 as first-line... Abstract Objective To compare efficacy and safety of nabpaclitaxel plus S-1 (AS) with gemcitabine plus S-1 (GS) as first-line treatment for metastatic pancreatic cancer. Methods We conducted a retrospective cohort, single-institution analysis by reviewing medical records of 38 patients who received either AS (nabpaclitaxel 125 mg/m2 on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14) or GS (gemcitabine 1000 mg/m2 on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14) chemotherapy. Results AS was associated with a prolonged median time to progression (TTP; 7.1 months, 95% CI, 4.5–9.7 vs. 3.6 months, 95% CI, 1.8–5.4; P value = 0.022) and improved median overall survival (OS; 10.2 months, 95% CI, 9.1–11.3 vs. 6 months, 95% CI, 4.2–7.8; P value <0.001) compared with GS. In cox proportional hazards model, treatment regimen was the only variable to be significantly associated with improvements in both TTP and OS. Subgroup analyses based on HER2 expression showed that AS seemed to have better outcome of OS in HER2 positive patients (HR = 0.168; 95% CI, 0.022–1.27; P value = 0.084). Hematological adverse events were commonly seen in both group (12.5% and 22.7%, GS and AS group, Grade 3 or 4; P value = 0.675) while AS got increased risk of sensory neuropathy (6 of 22 patients in AS, 27.3% vs. 0 of 16 patients in GS, all grade; P value = 0.03). Conclusions AS could be an effective treatment regimen for metastatic pancreatic cancer under surveillance of toxicity. nabpaclitaxel, gemcitabine, S-1, metastatic pancreatic cancer, HER2 Introduction Pancreatic ductal adenocarcinoma (PDAC) is the seventh leading cause of cancer-related death in the world (1). In 2015, there were 90 100 newly diagnosed patients, while 79 400 patients were died of pancreatic cancer in China (2). Surgical resection is the only potential curative care. Unfortunately, because of late presentation of relative symptoms, only 15–20% of patients are eligible for surgery. For locally advanced and metastatic pancreatic cancer, Fluorouracil (FU)-based chemotherapy was the mainstay of treatment since the 1950s (3). Gemcitabine was the first chemotherapeutic drug shown to be superior to 5-FU in terms of increasing overall survival (OS), but the prognosis remains grim (4). Thus, researchers were eager to explore more effective combination regimen, either gemcitabine or FU-based. The superiority of short-term infusional FU, leucovorin, oxaliplatin and irinotecan (FOLFIRINOX) over gemcitabine monotherapy was suggested in randomized Phase II and then Phase III clinical trial. However, treatment-related toxicity was also significantly worse with FOLFIRINOX (5,6). For gemcitabine-based regimen, at least four randomized clinical trials have directly compared gemcitabine plus S-1 with gemcitabine alone. All have shown that combination therapy significantly improved progression free survival (PFS) and objective response rate (ORR) but also associated with greater treatment-related toxicity. Besides, three of the four did not show a survival advantage to combined therapy (7–11). The multi-national MPACT trial recruited 861 patients with previously untreated metastatic pancreatic adenocarcinoma. The trial showed that combination of nabpaclitaxel (Abraxane) and gemcitabine, compared with gemcitabine alone, was associated with significantly higher ORR (23 vs. 7%), and significantly longer OS (8.5 vs. 6.7 months) and PFS (5.5 vs. 3.7 months) while Grade 3 or 4 adverse events were seen more often with combination therapy (12). In GEST and JASPAC01, rate of chemotherapy-related adverse events especially hematologic toxicity is lower in S-1 monotherapy comparing with gemcitabine alone (10,13). Therefore, we wonder whether S-1 will be a better alternative combined with nabpaclitaxel. In preclinical model, the researches of our team and Suenaga et al. both showed that S-1 plus nabpaclitaxel exerted better antitumor activity than that of the single agent (14,15). Meanwhile, we found that efficacy of nabpaclitaxel was better in HER2-positive xenograft mouse model (14). Besides, a recent published Phase II clinical trial demonstrated that S-1 plus nabpaclitaxel got encouraging ORR and manageable toxicities (16). In the present study, we compared the efficacy and safety of first-line therapy with either gemcitabine plus S-1 (GS) or nabpaclitaxel plus S-1 (AS) for metastatic pancreatic cancer. Additionally, we conducted a subgroup analysis to explore the correlation of HER2 expression with time to progression (TTP) and OS. Materials and methods Patients selection We retrospectively analyzed medical records of patients with metastatic pancreatic adenocarcinoma treated with either GS or AS as first-line therapy between April 2016 and December 2017 in Zhongshan Hospital, Fudan University. Eligible patients were required to have histologically or cytologically proven metastatic pancreatic adenocarcinoma. This study was approved by the ethical committee of our hospital and all patients have signed informed consent forms before we collected their medical records for researching purpose. Treatment option We would recommend nabpaclitaxel plus S-1 as frontline therapy because previous clinical trails already showed that both S-1 and nabpaclitaxel plus gemcitabine were associated with improved OS and less toxicity compared with gemcitabine alone (10,17,18). Since nabpaclitaxel was quite expensive and was not covered by the medical insurance in China, gemcitabine plus S-1 would be an alternative for patients who could not afford nabpaclitaxel. Study design This study was a retrospective cohort, single-institution analysis. Its objectives were to compare the efficacy and safety of GS with that of AS as first-line treatment for metastatic pancreatic adenocarcinoma. The response rate was defined according to revised Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Either computed tomography or magnetic resonance imaging was performed at baseline and then every three cycles until disease progression. Evaluations of treatment response according to the radiology reports were done by the same oncologist in a blinded manner. Toxic effects were graded using the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. HER2 expression was detected via immunohistochemistry of biopsy harvested tumor sample. The slides were reviewed by an pathologist in a blinded manner. The HER2 level were assessed by the immunoreactive score, which was calculated as the product of staining intensity (on a scale of 0–3) and percentage of positive cells (on a scale of 1–5) (19). Sections with a score of more than three were classified as positive; otherwise, negative. Because of the retrospective nature of our study, no quality of life questionnaire was completed. Study treatment GS consisted of 1000 mg/m2 gemcitabine over 30 min on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14. AS consisted of 125 mg/m2 nabpaclitaxel over 30 min on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14. Both regimens were repeated every 3 weeks until tumor progression or intolerable toxicity occurred. Dose reductions or regimen adjustment were planned according to the severity of chemotherapy-related toxicity. Second-line treatment was not defined by protocol and was at investigator’s discretion. Data analysis We tested normality and homogeneity of variance using Shapiro–Wilk test and Levene’s test. Patient age, baseline serum albumin, baseline CA125, maximum standardized uptake value (SUVmax) of pancreatic lesion on PET/CT, median total dose of S-1 and median dose intensity of S-1 were compared using independent sample t-test. Baseline C-reactive protein, baseline CA199, baseline squamous cell carcinoma antigen (SCC) and SUVmax of hepatic lesion were compared using Mann–Whitney U test. Sex, ECOG performance status (ECOG/PS), primary lesion location and metastatic sites were compared using Fisher’s exact test. TTP was measured from the first day of chemotherapy to the time of disease progression. OS was measured from the first day of chemotherapy to the time of death or last follow-up. The period of follow-up was estimated using the reverse Kaplan–Meier method. OS and TTP were calculated using the Kaplan–Meier method and were compared using the log-rank or Breslow test. The hazard ratio (HR) was calculated using univariate and multiple Cox proportional hazard regression modeling. Likelihood ratio tests of HER2 expression according to covariate interactions were used to examine the heterogeneity of the treatment effect according to prespecified subgroups of the treatment regimen. Statistical analyses were carried out using IBM SPSS Statistics version 22. Response evaluation based on RECIST version 1.1 was compared using Fisher’s exact test. Adverse events were classified according to CTCAE version 4.0. and compared with Fisher’s exact test. Results Patients characteristics We retrospectively reviewed the medical record of 16 patients treated with GS and 22 with AS as first-line chemotherapy for metastatic pancreatic cancer in our institution between April 2016 and December 2017. Baseline characteristics were well balanced between the two groups. The median age was 62.6 years in the GS group and 62.1 years in the AS group. All patients were assessed as ECOG/PS 0 or 1 in both groups, except for one patient was assessed as ECOG/PS 2 in the AS group. Other patient characteristics are listed in Table 1. Table 1. Characteristic of the patients at baseline Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) GS, gemcitabine plus S-1; AS, nabpaclitaxel plus S-1; SUVmax, maximum standardized uptake value of PET/CT. aMann–Whitney U test. bindependent sample t-test. cFisher’s exact test. Table 1. Characteristic of the patients at baseline Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) GS, gemcitabine plus S-1; AS, nabpaclitaxel plus S-1; SUVmax, maximum standardized uptake value of PET/CT. aMann–Whitney U test. bindependent sample t-test. cFisher’s exact test. Treatment compliance The median treatment cycles were 3.5 (range, 1–7, total 58 cycles) and 4 (range, 1–12, total 114 cycles), in GS and AS group, respectively. Median total dose of S-1 was 225 mg/m2 in GS group (range, 62.5–480 mg/m2) and 250 mg/m2 in AS group (range, 75–600 mg/m2; P = 0.404, independent sample t-test). Median dose intensity of S-1 was 71.4 mg/m2/d (range, 62.5–80 mg/m2/d) and 63.16 mg/m2/d (range, 18.75–85.71 mg/m2/d) on Days 1 through 14, corresponding to 89.25% and 78.95% of standard dose intensity, in GS and AS, respectively (P = 0.03, independent sample t-test). In GS group, median total dose of gemcitabine was 6000 mg/m2 (range, 1740–13 300 mg/m2) and median dose intensity of gemcitabine was 647.7 mg/m2/week (range, 533.3–933.3 mg/m2/week), corresponding to 97.2% of the planned dose intensity. In AS group, median total dose of nabpaclitaxel was 1000 mg/m2 (range, 256.4–3000 mg/m2) and median dose intensity was 82.2 mg/m2/week (range, 70.2–95.2 mg/m2/week), corresponding to 98.6% of the planned dose intensity. In GS group, treatment continues as planned in 5 of 16 patients. Two patients had dose reduction because of chemotherapy-related toxicity. The rest of GS group patients discontinued treatment due to disease progression. In AS group, 10 of 22 patients still receive planned treatment. Three patients had treatment response but discontinued chemotherapy due to patients’ intention. Totally eight patients got revised regimen because of toxicity. Three of them received nabpaclitaxel monotherapy while the other three patients were treated with S-1 alone. One patient had reduction in S-1 dose and 1 patient replaced S-1 with capecitabine. The rest of AS group patients discontinued treatment due to disease progression. Efficacy Four patients have not yet been evaluated for treatment response when review was conducted. The ORR was 6.7% (1 partial response [PR] of 16 patients) in GS group and 42.1% (8 PR of 22 patients) in AS group (P value = 0.047). The disease control rate (DCR) was 60% (1 PR and eight stable disease [SD] of 16 patients) in GS group and 84% (8 PR and 8 SD of 22 patients) in AS group (P value = 0.139). Median TTP was 3.6 (95% CI, 1.8–5.4) months for patients treated with GS compared with 7.1 (95% CI, 4.5–9.7) months for patients prescribed with AS (P value = 0.022). Median OS was 6 (95% CI, 4.2–7.8) months in GS group compared with 10.2 (95% CI, 9.1–11.3) months for patients in AS group (P value <0.001). Efficacy of both groups TTP and OS curves are summarized in Fig. 1. Figure 1. View largeDownload slide (A) Efficacy summary of GS and AS. *, Fisher’s exact test; †, Breslow test, ‡, Log-rank test. PR, partial response; SD, stable disease; PD, progression disease; DCR, disease control rate; ORR, objective response rate; TTP, time to progression; OS, overall survival. (B) Kaplan–Meier estimates of progression free survival and overall survival. Median TTP was 3.6 (95% CI, 1.8–5.4) in GS group versus 7.1 (95% CI, 4.5–9.7) months in AS group. Median OS was 6 (95% CI, 4.2–7.8) months in GS group versus 10.2 (95% CI, 9.1–11.3) months in AS group. CI, confidence interval. Figure 1. View largeDownload slide (A) Efficacy summary of GS and AS. *, Fisher’s exact test; †, Breslow test, ‡, Log-rank test. PR, partial response; SD, stable disease; PD, progression disease; DCR, disease control rate; ORR, objective response rate; TTP, time to progression; OS, overall survival. (B) Kaplan–Meier estimates of progression free survival and overall survival. Median TTP was 3.6 (95% CI, 1.8–5.4) in GS group versus 7.1 (95% CI, 4.5–9.7) months in AS group. Median OS was 6 (95% CI, 4.2–7.8) months in GS group versus 10.2 (95% CI, 9.1–11.3) months in AS group. CI, confidence interval. Univariate and multivariate Cox regression analysis of TTP and OS Cox regression analysis for TTP was performed to identify potential prognostic factors associated with patient outcome. Because of small sample size, we assumed variable to be statistically significant when P value was lower than 0.2 in univariate model. We found that treatment regimen, age, metastatic sites, HER2 expression of liver or pancreas, and SUVmax of pancreatic lesion were associated with TTP (P value = 0.017, 0.143, 0.131, 0.144 and 0.159, respectively). We then tested all these variables with multivariate Cox regression model. Only treatment regimen remained significant (AS to GS, HR = 0.032, 95% CI = 0.002–0.669, P value = 0.026). We further performed univariate and multivariate Cox analysis with OS as the outcome. Treatment regimen and HER2 expression of liver or pancreas were identified to have a statistically significant association with OS in univariate model (P value = 0.002 and 0.032, respectively), but only treatment regimen remained to be significant in multivariate model (AS to GS, HR = 0.172, 95% CI = 0.046–0.639, P value = 0.009). Other data are listed in Table 2. Table 2. Cox regression analysis for TTP and OS Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 TTP, time to progression; OS, overall survival. aWe assumed variable to be statistically significant when P value was lower than 0.2 on univariate analysis because of small sample size. bUnivariate Cox regression analysis. cMultivariate Cox regression analysis. Table 2. Cox regression analysis for TTP and OS Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 TTP, time to progression; OS, overall survival. aWe assumed variable to be statistically significant when P value was lower than 0.2 on univariate analysis because of small sample size. bUnivariate Cox regression analysis. cMultivariate Cox regression analysis. Subgroup analysis Exploratory analysis by subgroups according to HER2 expression of liver or pancreas is shown in Fig. 2. In the prespecified subgroup analysis, no significant TTP and OS difference were detected. The tests for interaction between treatment regimen and HER2 expression were 0.013 and 0.003 within TTP and OS subgroup analysis respectively. Figure 2. View largeDownload slide Treatment effect on time to progression and overall survival in prespecified subgroups. Hazard ratios are shown on a logarithmic scale. Figure 2. View largeDownload slide Treatment effect on time to progression and overall survival in prespecified subgroups. Hazard ratios are shown on a logarithmic scale. Toxicity All patients who received at least one cycle of chemotherapy were considered eligible for toxicity analysis (n = 38). The frequencies of adverse events in both groups are summarized in Table 3. Table 3. Comparison of safety profile between GS and AS Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – aFisher’s exact test. Table 3. Comparison of safety profile between GS and AS Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – aFisher’s exact test. No patients died from treatment-related cause in either group. The most common treatment-related adverse event was hematological adverse events in both group (31.3% and 45.5%, all grade; 12.5% and 22.7%, Grade 3 or 4, in GS and AS group, respectively) including anemia, neutropenia, and thrombocytopenia but remained statistically insignificant (P value = 0.506, all grade; P value = 0.675, Grade 3 or 4). However, we observed increased risk of sensory neuropathy in AS group (6 of 22 patients, 27.3%) while no patients experienced neurotoxicity in GS group (P value = 0.03, all grade). In addition, one patient was diagnosed with live abscess after two cycles of AS therapy. Discussion We conducted a retrospective analysis of metastatic pancreatic cancer patients who received GS or AS as first-line chemotherapy in our institution. When comparing efficacy of GS group, we observed increased ORR (42.1–6.7%, P value = 0.047), TTP (7.1–3.6 months, P value = 0.022), and OS (10.2–6 months, P value <0.001) in AS group. The Cox regression analysis showed that treatment regimen was the only variable significantly associated with TTP (AS to GS, HR = 0.032, 95% CI = 0.002–0.669, P value = 0.026) and OS (AS to GS, HR = 0.172, 95% CI = 0.046–0.639, P value = 0.009). While efficacy of AS in the present study was similar to that of a recent Phase II clinical trial whose median PFS and OS were 5.6 and 9.4 month, respectively (16), GS group seemed to have worse outcome compared with previous clinical trials especially worse OS (6 vs. 8.6–13.7 months) (7–10). Despite small simple size, this may be due to low socioeconomic status and treatment compliance of patients in GS group. Nabpaclitaxel was not covered by the medical insurance in China as I mentioned above and GS was an alternative for patients who could not afford AS. Some patients of GS group neither switch to second-line chemotherapy nor seek palliative care once progression occurred. Suenaga et al. and our previous research both showed that nabpaclitaxel brought about significant stroma reduction and enhanced vascularization in the preclinical model (14,15). These were similar to Von Hoff’s preclinical model of metastatic pancreatic cancer treated with nabpaclitaxel plus gemcitabine. They assumed that nabpaclitaxel reduced the dense of tumor stroma and may increase intratumoral concentration of gemcitabine (17). Nabpaclitaxel may also facilitate the entrance of S-1 into tumor tissue more efficiently. Our previous research also implied that efficacy of nabpaclitaxel was better in HER2-positive xenograft mouse model (14). Thus, we performed exploratory analyses of TTP and OS by subgroups according to HER2 expression of liver or pancreas. No significant TTP and OS difference were identified. But we still observed the trend that AS seemed to have better outcome of OS in HER2 positive patients (HER2 positive to negative, HR = 0.168, 95% CI = 0.022–1.27, P value = 0.084). This result may be due to the small sample size, but also emphasize the necessity for further investigation. Besides, previous study suggested that nabpaclitaxel was equal to or even better than polysorbate-based docetaxel in HER2 positive cell line PC3 and HT29 with medium to high albumin-binding protein SPARC (secreted protein, acidic and rich in cysteine) levels (20). We shall, therefore, investigate SPARC expression in pancreatic or hepatic lesion for further research. Combination of gemcitabine with either S-1 or nabpaclitaxel was related with more frequent chemotherapy-related adverse effect compared with gemcitabine alone in previous clinical trials (8–10,12). However, S-1 monotherapy showed less toxicity, especially hematological toxicity, when compared with gemcitabine (10,13). Besides, the recent Phase II clinical trial demonstrated manageable adverse events when patients with metastatic pancreatic adenocarcinoma were treated with AS (16). In the present study, we found increased risk of sensory neuropathy (27.3% to 0, P value = 0.03) and increased rate of hematological toxicity (45.5–31.3%) although it was not statistically significant (P value = 0.506). However, we still suggested that complete blood count test and sensory symptoms such as feelings of numbness should be closely monitored for both groups. AS does yield a better clinical outcome for metastatic pancreatic cancer patients than GS in our study, even though we must point out that this retrospective analysis was based on quite a small sample in one institution. These results should be confirmed by additional, prospective randomized trials. Funding This study was supported by National Natural Science Foundation of China (81773068). Conflict of interest statement We declare that they have no conflict of interest. References 1 Torre LA , Bray F , Siegel RL , Ferlay J , Lortet-Tieulent J , Jemal A . Global cancer statistics, 2012 . CA Cancer J Clin 2015 ; 65 : 87 – 108 . Google Scholar CrossRef Search ADS PubMed 2 Chen W , Zheng R , Baade PD , et al. . Cancer statistics in China, 2015 . CA Cancer J Clin 2016 ; 66 : 115 – 32 . Google Scholar CrossRef Search ADS PubMed 3 Moertel CG . Chemotherapy of gastrointestinal cancer . N Engl J Med 1978 ; 299 : 1049 – 52 . Google Scholar CrossRef Search ADS PubMed 4 Burris HR , Moore MJ , Andersen J , et al. . Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial . J Clin Oncol 1997 ; 15 : 2403 – 13 . Google Scholar CrossRef Search ADS PubMed 5 Ychou M , Desseigne F , Guimbaud R Randomized phase II trial comparing folfirinox (5FU/leucovorin [LV], irinotecan and oxaliplatin [O]) vs gemcitabine (G) as first-line treatment for metastatic pancreatic adenocarcinoma (MPA). First results of the ACCORD 11 trial (abstract 4516). J Clin Oncol. 6 Conroy T , Desseigne F , Ychou M , et al. . FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer . N Engl J Med 2011 ; 364 : 1817 – 25 . Google Scholar CrossRef Search ADS PubMed 7 Nakai Y , Isayama H , Sasaki T , et al. . A multicentre randomised phase II trial of gemcitabine alone vs gemcitabine and S-1 combination therapy in advanced pancreatic cancer: GEMSAP study . Br J Cancer 2012 ; 106 : 1934 – 9 . Google Scholar CrossRef Search ADS PubMed 8 Sudo K , Ishihara T , Hirata N , et al. . Randomized controlled study of gemcitabine plus S-1 combination chemotherapy versus gemcitabine for unresectable pancreatic cancer . Cancer Chemother Pharmacol 2014 ; 73 : 389 – 96 . Google Scholar CrossRef Search ADS PubMed 9 Ozaka M , Matsumura Y , Ishii H , et al. . Randomized phase II study of gemcitabine and S-1 combination versus gemcitabine alone in the treatment of unresectable advanced pancreatic cancer (Japan Clinical Cancer Research Organization PC-01 study) . Cancer Chemother Pharmacol 2012 ; 69 : 1197 – 204 . Google Scholar CrossRef Search ADS PubMed 10 Ueno H , Ioka T , Ikeda M , et al. . Randomized phase III study of gemcitabine plus S-1, S-1 alone, or gemcitabine alone in patients with locally advanced and metastatic pancreatic cancer in Japan and Taiwan: GEST study . J Clin Oncol 2013 ; 31 : 1640 – 8 . Google Scholar CrossRef Search ADS PubMed 11 Okusaka T , Miyakawa H , Fujii H , et al. . Updated results from GEST study: a randomized, three-arm phase III study for advanced pancreatic cancer . J Cancer Res Clin Oncol 2017 ; 143 : 1053 – 9 . Google Scholar CrossRef Search ADS PubMed 12 Von Hoff DD , Ervin T , Arena FP , et al. . Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine . N Engl J Med 2013 ; 369 : 1691 – 703 . Google Scholar CrossRef Search ADS PubMed 13 Maeda A , Boku N , Fukutomi A , et al. . Randomized phase III trial of adjuvant chemotherapy with gemcitabine versus S-1 in patients with resected pancreatic cancer: Japan Adjuvant Study Group of Pancreatic Cancer (JASPAC-01) . Jpn J Clin Oncol 2008 ; 38 : 227 – 9 . Google Scholar CrossRef Search ADS PubMed 14 Li JA , Xu XF , Han X , et al. . Nab-Paclitaxel Plus S-1 shows increased antitumor activity in patient-derived pancreatic cancer xenograft mouse models . Pancreas 2016 ; 45 : 425 – 33 . Google Scholar CrossRef Search ADS PubMed 15 Suenaga M , Yamada S , Fujii T , et al. . S-1 plus nab-paclitaxel is a promising regimen for pancreatic cancer in a preclinical model . J Surg Oncol 2016 ; 113 : 413 – 9 . Google Scholar CrossRef Search ADS PubMed 16 Shi Y , Zhang S , Han Q , et al. . Nab-paclitaxel plus S-1 in advanced pancreatic adenocarcinoma (NPSPAC): a single arm, single center, phase II trial . Oncotarget 2017 ; 8 : 92401 – 10 . Google Scholar PubMed 17 Von Hoff DD , Ramanathan RK , Borad MJ , et al. . Gemcitabine plus nab-paclitaxel is an active regimen in patients with advanced pancreatic cancer: a phase I/II trial . J Clin Oncol 2011 ; 29 : 4548 – 54 . Google Scholar CrossRef Search ADS PubMed 18 Von Hoff DD , Ervin T , Arena FP , et al. . Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine . N Engl J Med 2013 ; 369 : 1691 – 703 . Google Scholar CrossRef Search ADS PubMed 19 Remmele W , Stegner HE . Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue . Pathologe 1987 ; 8 : 138 – 40 . Google Scholar PubMed 20 Desai NP , Trieu V , Hwang LY , Wu R , Soon-Shiong P , Gradishar WJ . Improved effectiveness of nanoparticle albumin-bound (nab) paclitaxel versus polysorbate-based docetaxel in multiple xenografts as a function of HER2 and SPARC status . Anticancer Drugs 2008 ; 19 : 899 – 909 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Japanese Journal of Clinical Oncology Oxford University Press

Efficacy and safety comparison of nabpaclitaxel plus S-1 and gemcitabine plus S-1 as first-line chemotherapy for metastatic pancreatic cancer

Loading next page...
 
/lp/ou_press/efficacy-and-safety-comparison-of-nabpaclitaxel-plus-s-1-and-jH8VkXPjge
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0368-2811
eISSN
1465-3621
D.O.I.
10.1093/jjco/hyy063
Publisher site
See Article on Publisher Site

Abstract

Abstract Objective To compare efficacy and safety of nabpaclitaxel plus S-1 (AS) with gemcitabine plus S-1 (GS) as first-line treatment for metastatic pancreatic cancer. Methods We conducted a retrospective cohort, single-institution analysis by reviewing medical records of 38 patients who received either AS (nabpaclitaxel 125 mg/m2 on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14) or GS (gemcitabine 1000 mg/m2 on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14) chemotherapy. Results AS was associated with a prolonged median time to progression (TTP; 7.1 months, 95% CI, 4.5–9.7 vs. 3.6 months, 95% CI, 1.8–5.4; P value = 0.022) and improved median overall survival (OS; 10.2 months, 95% CI, 9.1–11.3 vs. 6 months, 95% CI, 4.2–7.8; P value <0.001) compared with GS. In cox proportional hazards model, treatment regimen was the only variable to be significantly associated with improvements in both TTP and OS. Subgroup analyses based on HER2 expression showed that AS seemed to have better outcome of OS in HER2 positive patients (HR = 0.168; 95% CI, 0.022–1.27; P value = 0.084). Hematological adverse events were commonly seen in both group (12.5% and 22.7%, GS and AS group, Grade 3 or 4; P value = 0.675) while AS got increased risk of sensory neuropathy (6 of 22 patients in AS, 27.3% vs. 0 of 16 patients in GS, all grade; P value = 0.03). Conclusions AS could be an effective treatment regimen for metastatic pancreatic cancer under surveillance of toxicity. nabpaclitaxel, gemcitabine, S-1, metastatic pancreatic cancer, HER2 Introduction Pancreatic ductal adenocarcinoma (PDAC) is the seventh leading cause of cancer-related death in the world (1). In 2015, there were 90 100 newly diagnosed patients, while 79 400 patients were died of pancreatic cancer in China (2). Surgical resection is the only potential curative care. Unfortunately, because of late presentation of relative symptoms, only 15–20% of patients are eligible for surgery. For locally advanced and metastatic pancreatic cancer, Fluorouracil (FU)-based chemotherapy was the mainstay of treatment since the 1950s (3). Gemcitabine was the first chemotherapeutic drug shown to be superior to 5-FU in terms of increasing overall survival (OS), but the prognosis remains grim (4). Thus, researchers were eager to explore more effective combination regimen, either gemcitabine or FU-based. The superiority of short-term infusional FU, leucovorin, oxaliplatin and irinotecan (FOLFIRINOX) over gemcitabine monotherapy was suggested in randomized Phase II and then Phase III clinical trial. However, treatment-related toxicity was also significantly worse with FOLFIRINOX (5,6). For gemcitabine-based regimen, at least four randomized clinical trials have directly compared gemcitabine plus S-1 with gemcitabine alone. All have shown that combination therapy significantly improved progression free survival (PFS) and objective response rate (ORR) but also associated with greater treatment-related toxicity. Besides, three of the four did not show a survival advantage to combined therapy (7–11). The multi-national MPACT trial recruited 861 patients with previously untreated metastatic pancreatic adenocarcinoma. The trial showed that combination of nabpaclitaxel (Abraxane) and gemcitabine, compared with gemcitabine alone, was associated with significantly higher ORR (23 vs. 7%), and significantly longer OS (8.5 vs. 6.7 months) and PFS (5.5 vs. 3.7 months) while Grade 3 or 4 adverse events were seen more often with combination therapy (12). In GEST and JASPAC01, rate of chemotherapy-related adverse events especially hematologic toxicity is lower in S-1 monotherapy comparing with gemcitabine alone (10,13). Therefore, we wonder whether S-1 will be a better alternative combined with nabpaclitaxel. In preclinical model, the researches of our team and Suenaga et al. both showed that S-1 plus nabpaclitaxel exerted better antitumor activity than that of the single agent (14,15). Meanwhile, we found that efficacy of nabpaclitaxel was better in HER2-positive xenograft mouse model (14). Besides, a recent published Phase II clinical trial demonstrated that S-1 plus nabpaclitaxel got encouraging ORR and manageable toxicities (16). In the present study, we compared the efficacy and safety of first-line therapy with either gemcitabine plus S-1 (GS) or nabpaclitaxel plus S-1 (AS) for metastatic pancreatic cancer. Additionally, we conducted a subgroup analysis to explore the correlation of HER2 expression with time to progression (TTP) and OS. Materials and methods Patients selection We retrospectively analyzed medical records of patients with metastatic pancreatic adenocarcinoma treated with either GS or AS as first-line therapy between April 2016 and December 2017 in Zhongshan Hospital, Fudan University. Eligible patients were required to have histologically or cytologically proven metastatic pancreatic adenocarcinoma. This study was approved by the ethical committee of our hospital and all patients have signed informed consent forms before we collected their medical records for researching purpose. Treatment option We would recommend nabpaclitaxel plus S-1 as frontline therapy because previous clinical trails already showed that both S-1 and nabpaclitaxel plus gemcitabine were associated with improved OS and less toxicity compared with gemcitabine alone (10,17,18). Since nabpaclitaxel was quite expensive and was not covered by the medical insurance in China, gemcitabine plus S-1 would be an alternative for patients who could not afford nabpaclitaxel. Study design This study was a retrospective cohort, single-institution analysis. Its objectives were to compare the efficacy and safety of GS with that of AS as first-line treatment for metastatic pancreatic adenocarcinoma. The response rate was defined according to revised Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. Either computed tomography or magnetic resonance imaging was performed at baseline and then every three cycles until disease progression. Evaluations of treatment response according to the radiology reports were done by the same oncologist in a blinded manner. Toxic effects were graded using the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0. HER2 expression was detected via immunohistochemistry of biopsy harvested tumor sample. The slides were reviewed by an pathologist in a blinded manner. The HER2 level were assessed by the immunoreactive score, which was calculated as the product of staining intensity (on a scale of 0–3) and percentage of positive cells (on a scale of 1–5) (19). Sections with a score of more than three were classified as positive; otherwise, negative. Because of the retrospective nature of our study, no quality of life questionnaire was completed. Study treatment GS consisted of 1000 mg/m2 gemcitabine over 30 min on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14. AS consisted of 125 mg/m2 nabpaclitaxel over 30 min on Days 1, 8 and S-1 80 mg/m2 on Days 1 to 14. Both regimens were repeated every 3 weeks until tumor progression or intolerable toxicity occurred. Dose reductions or regimen adjustment were planned according to the severity of chemotherapy-related toxicity. Second-line treatment was not defined by protocol and was at investigator’s discretion. Data analysis We tested normality and homogeneity of variance using Shapiro–Wilk test and Levene’s test. Patient age, baseline serum albumin, baseline CA125, maximum standardized uptake value (SUVmax) of pancreatic lesion on PET/CT, median total dose of S-1 and median dose intensity of S-1 were compared using independent sample t-test. Baseline C-reactive protein, baseline CA199, baseline squamous cell carcinoma antigen (SCC) and SUVmax of hepatic lesion were compared using Mann–Whitney U test. Sex, ECOG performance status (ECOG/PS), primary lesion location and metastatic sites were compared using Fisher’s exact test. TTP was measured from the first day of chemotherapy to the time of disease progression. OS was measured from the first day of chemotherapy to the time of death or last follow-up. The period of follow-up was estimated using the reverse Kaplan–Meier method. OS and TTP were calculated using the Kaplan–Meier method and were compared using the log-rank or Breslow test. The hazard ratio (HR) was calculated using univariate and multiple Cox proportional hazard regression modeling. Likelihood ratio tests of HER2 expression according to covariate interactions were used to examine the heterogeneity of the treatment effect according to prespecified subgroups of the treatment regimen. Statistical analyses were carried out using IBM SPSS Statistics version 22. Response evaluation based on RECIST version 1.1 was compared using Fisher’s exact test. Adverse events were classified according to CTCAE version 4.0. and compared with Fisher’s exact test. Results Patients characteristics We retrospectively reviewed the medical record of 16 patients treated with GS and 22 with AS as first-line chemotherapy for metastatic pancreatic cancer in our institution between April 2016 and December 2017. Baseline characteristics were well balanced between the two groups. The median age was 62.6 years in the GS group and 62.1 years in the AS group. All patients were assessed as ECOG/PS 0 or 1 in both groups, except for one patient was assessed as ECOG/PS 2 in the AS group. Other patient characteristics are listed in Table 1. Table 1. Characteristic of the patients at baseline Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) GS, gemcitabine plus S-1; AS, nabpaclitaxel plus S-1; SUVmax, maximum standardized uptake value of PET/CT. aMann–Whitney U test. bindependent sample t-test. cFisher’s exact test. Table 1. Characteristic of the patients at baseline Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) Characteristics Total (n = 38) GS (n = 16) AS (n = 22) P value Age 0.731b  Median 62.6 [46–76] 62.1 [49–71] 63.0 [46–76] Gender (%) 1c  Male 27 10 (71.4) 17 (70.8)  Female 11 4 (28.6) 7 (28.2) ECOG/PS (%) 0.68c  0 5 2 (12.5) 3 (13.6)  1 32 14 (87.5) 18 (81.8)  2 1 0 (0) 1 (4.5) Baseline albumin level (g/l) 0.318b  Median 43.5 (33–51) 44.2 (40–51) 43.0 (33–51) Baseline CRP level (mg/l) 0.934a  Median 21.4 (0–200) 11.3 (0.7–40) 28.7 (0–200) Baseline CA199 level (U/ml) 0.495a  Median 3442.0 (0.7–>10 000) 3599.2 (0.7–>10 000) 3327.8 (40–>10 000) Baseline CA125 level (U/ml) 0.186b  Median 404.3 (0–4823) 159.1 (13–1344) 582.6 (0–4823) Baseline SCC level (ng/ml) 0.658a  Median 2.1 (0–33.5) 0.8 (0.4–1.5) 3.0 (0–33.5) Primary tumor location on pancreas (%) 0.101c  Head 13 8 (50) 5 (22.7)  Neck 4 3 (18.8) 1 (4.5)  Body 5 1 (6.3) 4 (18.2)  Tail 9 1 (6.3) 8 (36.4)  Body + Tail 6 3 (18.8) 3 (13.6)  Whole pancreas 1 0 (0) 1 (4.5) Metastatic site (%) 0.427c  Liver 17 7 (43.8) 10 (45.5)  Liver and other sites 13 7 (43.8) 6 (27.3)  Except for liver 8 2 (12.4) 6 (27.3) HER2 expression in liver or pancreas 0.264c  Positive 18 6 (46.2) 12 (70.6)  Negative 12 7 (53.8) 5 (29.4) SUVmax of pancreatic lesion 0.936b  Median 6.9 (0–15.8) 6.9 (3.9–12.1) 6.8 (0–15.8) SUVmax of hepatic lesion 0.556a  Median 4.3 (2.2–5.9) 4.8 (3.7–5.9) 4.0 (2.2–4.9) GS, gemcitabine plus S-1; AS, nabpaclitaxel plus S-1; SUVmax, maximum standardized uptake value of PET/CT. aMann–Whitney U test. bindependent sample t-test. cFisher’s exact test. Treatment compliance The median treatment cycles were 3.5 (range, 1–7, total 58 cycles) and 4 (range, 1–12, total 114 cycles), in GS and AS group, respectively. Median total dose of S-1 was 225 mg/m2 in GS group (range, 62.5–480 mg/m2) and 250 mg/m2 in AS group (range, 75–600 mg/m2; P = 0.404, independent sample t-test). Median dose intensity of S-1 was 71.4 mg/m2/d (range, 62.5–80 mg/m2/d) and 63.16 mg/m2/d (range, 18.75–85.71 mg/m2/d) on Days 1 through 14, corresponding to 89.25% and 78.95% of standard dose intensity, in GS and AS, respectively (P = 0.03, independent sample t-test). In GS group, median total dose of gemcitabine was 6000 mg/m2 (range, 1740–13 300 mg/m2) and median dose intensity of gemcitabine was 647.7 mg/m2/week (range, 533.3–933.3 mg/m2/week), corresponding to 97.2% of the planned dose intensity. In AS group, median total dose of nabpaclitaxel was 1000 mg/m2 (range, 256.4–3000 mg/m2) and median dose intensity was 82.2 mg/m2/week (range, 70.2–95.2 mg/m2/week), corresponding to 98.6% of the planned dose intensity. In GS group, treatment continues as planned in 5 of 16 patients. Two patients had dose reduction because of chemotherapy-related toxicity. The rest of GS group patients discontinued treatment due to disease progression. In AS group, 10 of 22 patients still receive planned treatment. Three patients had treatment response but discontinued chemotherapy due to patients’ intention. Totally eight patients got revised regimen because of toxicity. Three of them received nabpaclitaxel monotherapy while the other three patients were treated with S-1 alone. One patient had reduction in S-1 dose and 1 patient replaced S-1 with capecitabine. The rest of AS group patients discontinued treatment due to disease progression. Efficacy Four patients have not yet been evaluated for treatment response when review was conducted. The ORR was 6.7% (1 partial response [PR] of 16 patients) in GS group and 42.1% (8 PR of 22 patients) in AS group (P value = 0.047). The disease control rate (DCR) was 60% (1 PR and eight stable disease [SD] of 16 patients) in GS group and 84% (8 PR and 8 SD of 22 patients) in AS group (P value = 0.139). Median TTP was 3.6 (95% CI, 1.8–5.4) months for patients treated with GS compared with 7.1 (95% CI, 4.5–9.7) months for patients prescribed with AS (P value = 0.022). Median OS was 6 (95% CI, 4.2–7.8) months in GS group compared with 10.2 (95% CI, 9.1–11.3) months for patients in AS group (P value <0.001). Efficacy of both groups TTP and OS curves are summarized in Fig. 1. Figure 1. View largeDownload slide (A) Efficacy summary of GS and AS. *, Fisher’s exact test; †, Breslow test, ‡, Log-rank test. PR, partial response; SD, stable disease; PD, progression disease; DCR, disease control rate; ORR, objective response rate; TTP, time to progression; OS, overall survival. (B) Kaplan–Meier estimates of progression free survival and overall survival. Median TTP was 3.6 (95% CI, 1.8–5.4) in GS group versus 7.1 (95% CI, 4.5–9.7) months in AS group. Median OS was 6 (95% CI, 4.2–7.8) months in GS group versus 10.2 (95% CI, 9.1–11.3) months in AS group. CI, confidence interval. Figure 1. View largeDownload slide (A) Efficacy summary of GS and AS. *, Fisher’s exact test; †, Breslow test, ‡, Log-rank test. PR, partial response; SD, stable disease; PD, progression disease; DCR, disease control rate; ORR, objective response rate; TTP, time to progression; OS, overall survival. (B) Kaplan–Meier estimates of progression free survival and overall survival. Median TTP was 3.6 (95% CI, 1.8–5.4) in GS group versus 7.1 (95% CI, 4.5–9.7) months in AS group. Median OS was 6 (95% CI, 4.2–7.8) months in GS group versus 10.2 (95% CI, 9.1–11.3) months in AS group. CI, confidence interval. Univariate and multivariate Cox regression analysis of TTP and OS Cox regression analysis for TTP was performed to identify potential prognostic factors associated with patient outcome. Because of small sample size, we assumed variable to be statistically significant when P value was lower than 0.2 in univariate model. We found that treatment regimen, age, metastatic sites, HER2 expression of liver or pancreas, and SUVmax of pancreatic lesion were associated with TTP (P value = 0.017, 0.143, 0.131, 0.144 and 0.159, respectively). We then tested all these variables with multivariate Cox regression model. Only treatment regimen remained significant (AS to GS, HR = 0.032, 95% CI = 0.002–0.669, P value = 0.026). We further performed univariate and multivariate Cox analysis with OS as the outcome. Treatment regimen and HER2 expression of liver or pancreas were identified to have a statistically significant association with OS in univariate model (P value = 0.002 and 0.032, respectively), but only treatment regimen remained to be significant in multivariate model (AS to GS, HR = 0.172, 95% CI = 0.046–0.639, P value = 0.009). Other data are listed in Table 2. Table 2. Cox regression analysis for TTP and OS Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 TTP, time to progression; OS, overall survival. aWe assumed variable to be statistically significant when P value was lower than 0.2 on univariate analysis because of small sample size. bUnivariate Cox regression analysis. cMultivariate Cox regression analysis. Table 2. Cox regression analysis for TTP and OS Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 Variable TTPb OSb TTPc OSc HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value HR (95% CI) P value Treatment  GS 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  AS 0.308 (0.117–0.812) 0.017a 0.14 (0.041–0.485) 0.002a 0.032 (0.002–0.669) 0.026 0.172 (0.046–0.639) 0.009 Sex  Male 1 (Reference) – 1 (Reference) –  Female 1.398 (0.561–3.483) 0.472 0.779 (0.248–2.445) 0.669 Age 0.95 (0.886–1.018) 0.143a 0.953 (0.881–1.031) 0.228 0.904 (0.667–1.226) 0.517 Metastatic site 0.131a 0.845 0.939  Liver 1 (Reference) – 1 (Reference) – 1 (Reference) –  Liver and other sites 0.776 (0.281–2.145) 0.625 1.288 (0.395–4.196) 0.675 0.607 (0.034–10.813) 0.734  Except for liver 2.552 (0.851–7.653) 0.095 1.438 (0.37–5.587) 0.6 0.840 (0.113–6.264) 0.865 Primary tumor location 0.953 0.495  Head 1 (Reference) – 1 (Reference) –  Neck 1.012 (0.207–4.946) 0.988 1.597 (0.308–8.288) 0.578  Body 0.717 (0.196–2.632) 0.617 0.162 (0.016–1.635) 0.123  Tail 0.769 (0.234–2.522) 0.664 1.149 (0.349–3.777) 0.819  Body + Tail 1.476 (0.423–5.145) 0.541 0.533 (0.091–3.134) 0.487  Whole pancreas 0 (0) 0.989 – – Baseline albumin level 0.966 (0.85–1.098) 0.599 0.952 (0.838–1.081) 0.448 Baseline CRP level 1.005 (0.997–1.013) 0.213 1.002 (0.994–1.01) 0.667 ECOG/PS 0.652 0.386  0 1 (Reference) 1 (Reference)  1 1.564 (0.516–4.743) 0.429 2.612 (0.563–12.121) 0.22  2 2.401 (0.254–22.665) 0.445 4.621 (0.383–55.678) Baseline CA199 level 1 (1) 0.463 1 (1) 0.609 HER2 expression in liver or pancreas  Negative 1 (Reference) – 1 (Reference) – 1 (Reference) – 1 (Reference) –  Positive 0.507 (0.204–1.262) 0.144a 0.287 (0.091–0.9) 0.032a 0.759 (0.104–5.568) 0.787 0.5 (0.145–1.727) 0.273 SUVmax of pancreatic lesion 1.113 (0.959–1.292) 0.159a 1.119 (0.866–1.446) 0.392 1.253 (0.913–1.719) 0.162 SUVmax of hepatic lesion 0.675 (0.1–4.543) 0.686 299.777 (0–1.931*1011) 0.582 TTP, time to progression; OS, overall survival. aWe assumed variable to be statistically significant when P value was lower than 0.2 on univariate analysis because of small sample size. bUnivariate Cox regression analysis. cMultivariate Cox regression analysis. Subgroup analysis Exploratory analysis by subgroups according to HER2 expression of liver or pancreas is shown in Fig. 2. In the prespecified subgroup analysis, no significant TTP and OS difference were detected. The tests for interaction between treatment regimen and HER2 expression were 0.013 and 0.003 within TTP and OS subgroup analysis respectively. Figure 2. View largeDownload slide Treatment effect on time to progression and overall survival in prespecified subgroups. Hazard ratios are shown on a logarithmic scale. Figure 2. View largeDownload slide Treatment effect on time to progression and overall survival in prespecified subgroups. Hazard ratios are shown on a logarithmic scale. Toxicity All patients who received at least one cycle of chemotherapy were considered eligible for toxicity analysis (n = 38). The frequencies of adverse events in both groups are summarized in Table 3. Table 3. Comparison of safety profile between GS and AS Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – aFisher’s exact test. Table 3. Comparison of safety profile between GS and AS Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – Adverse events CTCAR v4.0 GS (n = 16) (%) AS (n = 22) (%) P valuea All grade Grade 3 or 4 All grade Grade 3 or 4 All grade Grade 3 or 4 Hematological adverse events 5 (31.3) 2 (12.5) 10 (45.5) 5 (22.7) 0.506 0.675 Non-hematological adverse events  Hand-foot syndrome 1 (6.3) 0 (0) 1 (4.5) 1 (4.5) 1 1  Sensory neuropathy 0 (0) 0 (0) 6 (27.3) 1 (4.5) 0.03 1  Rash 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  AST/ALT elevated 1 (6.3) 0 (0) 0 (0) 0 (0) 0.421 –  Oral mucositis 0 (0) 0 (0) 2 (9.1) 0 (0) 0.499 –  Anorexia 0 (0) 0 (0) 1 (4.5) 0 (0) 1 – aFisher’s exact test. No patients died from treatment-related cause in either group. The most common treatment-related adverse event was hematological adverse events in both group (31.3% and 45.5%, all grade; 12.5% and 22.7%, Grade 3 or 4, in GS and AS group, respectively) including anemia, neutropenia, and thrombocytopenia but remained statistically insignificant (P value = 0.506, all grade; P value = 0.675, Grade 3 or 4). However, we observed increased risk of sensory neuropathy in AS group (6 of 22 patients, 27.3%) while no patients experienced neurotoxicity in GS group (P value = 0.03, all grade). In addition, one patient was diagnosed with live abscess after two cycles of AS therapy. Discussion We conducted a retrospective analysis of metastatic pancreatic cancer patients who received GS or AS as first-line chemotherapy in our institution. When comparing efficacy of GS group, we observed increased ORR (42.1–6.7%, P value = 0.047), TTP (7.1–3.6 months, P value = 0.022), and OS (10.2–6 months, P value <0.001) in AS group. The Cox regression analysis showed that treatment regimen was the only variable significantly associated with TTP (AS to GS, HR = 0.032, 95% CI = 0.002–0.669, P value = 0.026) and OS (AS to GS, HR = 0.172, 95% CI = 0.046–0.639, P value = 0.009). While efficacy of AS in the present study was similar to that of a recent Phase II clinical trial whose median PFS and OS were 5.6 and 9.4 month, respectively (16), GS group seemed to have worse outcome compared with previous clinical trials especially worse OS (6 vs. 8.6–13.7 months) (7–10). Despite small simple size, this may be due to low socioeconomic status and treatment compliance of patients in GS group. Nabpaclitaxel was not covered by the medical insurance in China as I mentioned above and GS was an alternative for patients who could not afford AS. Some patients of GS group neither switch to second-line chemotherapy nor seek palliative care once progression occurred. Suenaga et al. and our previous research both showed that nabpaclitaxel brought about significant stroma reduction and enhanced vascularization in the preclinical model (14,15). These were similar to Von Hoff’s preclinical model of metastatic pancreatic cancer treated with nabpaclitaxel plus gemcitabine. They assumed that nabpaclitaxel reduced the dense of tumor stroma and may increase intratumoral concentration of gemcitabine (17). Nabpaclitaxel may also facilitate the entrance of S-1 into tumor tissue more efficiently. Our previous research also implied that efficacy of nabpaclitaxel was better in HER2-positive xenograft mouse model (14). Thus, we performed exploratory analyses of TTP and OS by subgroups according to HER2 expression of liver or pancreas. No significant TTP and OS difference were identified. But we still observed the trend that AS seemed to have better outcome of OS in HER2 positive patients (HER2 positive to negative, HR = 0.168, 95% CI = 0.022–1.27, P value = 0.084). This result may be due to the small sample size, but also emphasize the necessity for further investigation. Besides, previous study suggested that nabpaclitaxel was equal to or even better than polysorbate-based docetaxel in HER2 positive cell line PC3 and HT29 with medium to high albumin-binding protein SPARC (secreted protein, acidic and rich in cysteine) levels (20). We shall, therefore, investigate SPARC expression in pancreatic or hepatic lesion for further research. Combination of gemcitabine with either S-1 or nabpaclitaxel was related with more frequent chemotherapy-related adverse effect compared with gemcitabine alone in previous clinical trials (8–10,12). However, S-1 monotherapy showed less toxicity, especially hematological toxicity, when compared with gemcitabine (10,13). Besides, the recent Phase II clinical trial demonstrated manageable adverse events when patients with metastatic pancreatic adenocarcinoma were treated with AS (16). In the present study, we found increased risk of sensory neuropathy (27.3% to 0, P value = 0.03) and increased rate of hematological toxicity (45.5–31.3%) although it was not statistically significant (P value = 0.506). However, we still suggested that complete blood count test and sensory symptoms such as feelings of numbness should be closely monitored for both groups. AS does yield a better clinical outcome for metastatic pancreatic cancer patients than GS in our study, even though we must point out that this retrospective analysis was based on quite a small sample in one institution. These results should be confirmed by additional, prospective randomized trials. Funding This study was supported by National Natural Science Foundation of China (81773068). Conflict of interest statement We declare that they have no conflict of interest. References 1 Torre LA , Bray F , Siegel RL , Ferlay J , Lortet-Tieulent J , Jemal A . Global cancer statistics, 2012 . CA Cancer J Clin 2015 ; 65 : 87 – 108 . Google Scholar CrossRef Search ADS PubMed 2 Chen W , Zheng R , Baade PD , et al. . Cancer statistics in China, 2015 . CA Cancer J Clin 2016 ; 66 : 115 – 32 . Google Scholar CrossRef Search ADS PubMed 3 Moertel CG . Chemotherapy of gastrointestinal cancer . N Engl J Med 1978 ; 299 : 1049 – 52 . Google Scholar CrossRef Search ADS PubMed 4 Burris HR , Moore MJ , Andersen J , et al. . Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial . J Clin Oncol 1997 ; 15 : 2403 – 13 . Google Scholar CrossRef Search ADS PubMed 5 Ychou M , Desseigne F , Guimbaud R Randomized phase II trial comparing folfirinox (5FU/leucovorin [LV], irinotecan and oxaliplatin [O]) vs gemcitabine (G) as first-line treatment for metastatic pancreatic adenocarcinoma (MPA). First results of the ACCORD 11 trial (abstract 4516). J Clin Oncol. 6 Conroy T , Desseigne F , Ychou M , et al. . FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer . N Engl J Med 2011 ; 364 : 1817 – 25 . Google Scholar CrossRef Search ADS PubMed 7 Nakai Y , Isayama H , Sasaki T , et al. . A multicentre randomised phase II trial of gemcitabine alone vs gemcitabine and S-1 combination therapy in advanced pancreatic cancer: GEMSAP study . Br J Cancer 2012 ; 106 : 1934 – 9 . Google Scholar CrossRef Search ADS PubMed 8 Sudo K , Ishihara T , Hirata N , et al. . Randomized controlled study of gemcitabine plus S-1 combination chemotherapy versus gemcitabine for unresectable pancreatic cancer . Cancer Chemother Pharmacol 2014 ; 73 : 389 – 96 . Google Scholar CrossRef Search ADS PubMed 9 Ozaka M , Matsumura Y , Ishii H , et al. . Randomized phase II study of gemcitabine and S-1 combination versus gemcitabine alone in the treatment of unresectable advanced pancreatic cancer (Japan Clinical Cancer Research Organization PC-01 study) . Cancer Chemother Pharmacol 2012 ; 69 : 1197 – 204 . Google Scholar CrossRef Search ADS PubMed 10 Ueno H , Ioka T , Ikeda M , et al. . Randomized phase III study of gemcitabine plus S-1, S-1 alone, or gemcitabine alone in patients with locally advanced and metastatic pancreatic cancer in Japan and Taiwan: GEST study . J Clin Oncol 2013 ; 31 : 1640 – 8 . Google Scholar CrossRef Search ADS PubMed 11 Okusaka T , Miyakawa H , Fujii H , et al. . Updated results from GEST study: a randomized, three-arm phase III study for advanced pancreatic cancer . J Cancer Res Clin Oncol 2017 ; 143 : 1053 – 9 . Google Scholar CrossRef Search ADS PubMed 12 Von Hoff DD , Ervin T , Arena FP , et al. . Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine . N Engl J Med 2013 ; 369 : 1691 – 703 . Google Scholar CrossRef Search ADS PubMed 13 Maeda A , Boku N , Fukutomi A , et al. . Randomized phase III trial of adjuvant chemotherapy with gemcitabine versus S-1 in patients with resected pancreatic cancer: Japan Adjuvant Study Group of Pancreatic Cancer (JASPAC-01) . Jpn J Clin Oncol 2008 ; 38 : 227 – 9 . Google Scholar CrossRef Search ADS PubMed 14 Li JA , Xu XF , Han X , et al. . Nab-Paclitaxel Plus S-1 shows increased antitumor activity in patient-derived pancreatic cancer xenograft mouse models . Pancreas 2016 ; 45 : 425 – 33 . Google Scholar CrossRef Search ADS PubMed 15 Suenaga M , Yamada S , Fujii T , et al. . S-1 plus nab-paclitaxel is a promising regimen for pancreatic cancer in a preclinical model . J Surg Oncol 2016 ; 113 : 413 – 9 . Google Scholar CrossRef Search ADS PubMed 16 Shi Y , Zhang S , Han Q , et al. . Nab-paclitaxel plus S-1 in advanced pancreatic adenocarcinoma (NPSPAC): a single arm, single center, phase II trial . Oncotarget 2017 ; 8 : 92401 – 10 . Google Scholar PubMed 17 Von Hoff DD , Ramanathan RK , Borad MJ , et al. . Gemcitabine plus nab-paclitaxel is an active regimen in patients with advanced pancreatic cancer: a phase I/II trial . J Clin Oncol 2011 ; 29 : 4548 – 54 . Google Scholar CrossRef Search ADS PubMed 18 Von Hoff DD , Ervin T , Arena FP , et al. . Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine . N Engl J Med 2013 ; 369 : 1691 – 703 . Google Scholar CrossRef Search ADS PubMed 19 Remmele W , Stegner HE . Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue . Pathologe 1987 ; 8 : 138 – 40 . Google Scholar PubMed 20 Desai NP , Trieu V , Hwang LY , Wu R , Soon-Shiong P , Gradishar WJ . Improved effectiveness of nanoparticle albumin-bound (nab) paclitaxel versus polysorbate-based docetaxel in multiple xenografts as a function of HER2 and SPARC status . Anticancer Drugs 2008 ; 19 : 899 – 909 . Google Scholar CrossRef Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Japanese Journal of Clinical OncologyOxford University Press

Published: Apr 28, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off