A prospective study of XELIRI plus bevacizumab as the first-line therapy in Japanese patients with unresectable or recurrent colorectal cancer (KSCC1101)

752

Background: XELIRI plus bevacizumab (Bev) might be one of the standard therapies for metastatic colorectal cancer (mCRC). However, there was no clinical practice data in Japan. This study was designed to evaluate the efficacy and safety of XELIRI plus Bev in clinical practice in Japanese mCRC patients (pts.). Methods: This was a multicenter, single-arm, open-label prospective study. The major inclusion criteria were previously untreated mCRC; presence of measurable lesions; age ≥20 years; ECOG performance status (PS) of 0–2; and adequate organ function. Pts received Bev 7.5 mg/kg day 1 and XELIRI (irinotecan 200 mg/m2 day 1 plus capecitabine 800 mg/m2 bid d1-14) q3w. This schedule was repeated until unacceptable toxicity or disease progression occurred. The primary endpoint was RECIST-confirmed objective response rate (ORR). A sample size of 35 was planned for a threshold ORR of 30% and expected value of 53%, with one-sided alpha of 0.05 and beta of approximately 0.2. Results: A total of 36 pts (male/female, 23/13; median age, 65.0 years (range 44.0-80.0); PS 0/1/2, 31/5/0) were enrolled in this study from July 2011 to September 2012. One patient did not fulfill the eligibility criteria and one withdrew the informed consent before the start of protocol treatment. Thirty-four pts were assessed for response; CR 0, PR 18, SD 9, PD 4, and NE 3. The confirmed ORR was 52.9% (90% CI, 37.7-67.8%). Median PFS was 9.6 months (95% CI, 5.1-11.1 months) and mOS was 23.0 months (95% CI, 11.3-NE months). mTTF was 5.3 months (95% CI, 3.2-8.3 months). The most common grade 3/4 adverse events were neutropenia 31.4%, anemia 20.0%, thrombocytopenia 5.7%, hyponatremia 20.0%, anorexia 20.0%, diarrhea 22.8%, fatigue 0.0%, hypertension 2.9%, hand-foot syndrome 0.0%, and bleeding 0.0%. Conclusions: First-line treatment of XELIRI + Bev showed a promising response rate and an acceptable tolerability profile in the clinical practice in Japanese mCRC pts. Clinical trial information: UMIN000006070.

Liver resectability following S-1+L-OHP with cetuximab as the first-line treatment of unresectable liver limited metastases from KRAS exon 2 wild-colorectal cancer in Japanese patients (KSCC 1002)

755

Background: There is no data concerning liver resectability following S-1+L-OHP (C-SOX) with cetuximab (Cmab) as the first-line treatment of unresectable liver limited metastases from KRAS exon 2 wild-colorectal cancer by prospective, multi-center study in Japan. The Kyushu Study group of Clinical Cancer (KSCC) conducted phase II trials in this setting. Methods: Eligibility criteria included unresectable liver limited metastases from histologically confirmed advanced colorectal cancer, ECOG PS 0-1 and adequate general condition. Patients (pts) received 4-6 cycles of C-SOX (oxaliplatin 130 mg/m2, day 1 followed by S-1 80 mg/m2 po [days 1-14] every 3 weeks) with weekly durable intravenous (DIV) Cmab administration at 400 mg/m2 (day 1) and 250 mg/m2/week (except day 1), followed by evaluating the liver resectability. Results: Of the 33 pts enrolled from March 2010 to July 2013; M/F, 20/13; median age, 64 years (range 48-83); ECOG PS 0/1, 27/6; primary lesion -/+, 18/15; number of liver metastasis 1-4/5 or more, 10/23; uni-/bi-lobular liver meta. 5/28; synchronous/metachronous 30/3. The number of FAS and SAS cases was 33. Response without confirmation for CR, PR, SD, PD and NE were 1 (3.0%), 20 (60.6%), 6 (18.2%), 3 (9.1%) and 3 (9.1%), respectively. An overall response rate was 63.6% (95% CI: 46.3–81.0%). The liver resectability for all time treatment was 16/33 (48.5%). The number of R0 cases was 13 pts (39.4%). Median progression free survival (liver resection was not censored and not event, events = relapse, progression, secondary cancer, and death) for 16 liver resection cases and 17 not-resection cases were 25.0 months (8.1-not reached), 4.5 months (3.0-10.8), respectively. Median overall survival (mOS) for all 33cases was 31.6 months (14.8, not reached). Median OS for 17 not-resection cases was 14.0 (6.4-not reached), and mOS for 16 liver resection cases was not reached and 3-year survival rate was 51.4%. Conclusions: C-SOX + Cmab regimen is safe and effective for unresectable liver limited metastases from KRAS exon 2 wild-colorectal cancer, and might be to lead the high liver resectability. Clinical trial information: UMIN000004331.

A prospective study of capecitabine plus modified cisplatin (mXP) as first-line therapy in Japanese patients with metastatic gastric cancer (KSCC1104)

129

Background: Capecitabine plus cisplatin (XP) is one of the standard therapies for metastatic gastric cancer (mGC). However there is no clinical practice date in Japan, and results from the ToGA and AVAGAST study suggested dose of cisplatin should be lower in Japanese. This study was designed to evaluate the efficacy and safety of mXP in the clinical practice in Japanese mGC patients (pts) including the elderly. Methods: The study design was multicenter, single-arm, open-label prospective study. The major inclusion criteria were previously untreated mGC; presence of measurable lesions; age 20 years; ECOG performance status (PS) 0–2; and adequate organ function. Pts received mXP (cisplatin 60 mg/m2 d1 plus capecitabine 1,000 mg/m2 bid d1-14) q3w. This schedule was repeated until unacceptable toxicity or disease progression occurred. The primary endpoint was RECIST-confirmed objective response rate (ORR). A sample size of 40 was planned for a threshold ORR of 30% and expected value of 50%, with one-sided alpha of 0.05 and beta of approximately 0.2. Results: Of the 42 pts (male/female, 34/8; median age, 63.5 years (range 50-81); PS 0/1/2, 34/8/0) enrolled from Nov 2011 to Oct 2013. One patient did not fulfill the eligibility criteria. Forty one pts were assessed for response; CR 2 pts, PR 16 pts, SD 14 pts, PD 8 pts, and NE 1 pts. The confirmed ORR was 43.9% (90% CI; 31.2-56.7%, 95% CI; 28.7-59.1%). Median PFS and mOS were 4.5 months (95% CI; 3.9 – 6.5M) and 11.4 months (95% CI; 7.7 – not reached). The most common grade 3/4 adverse events were anorexia 24.4%, neutropenia 36.6%, fatigue 9.8%, hand-foot syndrome 7.3%. Grade 3/4 peripheral neuropathy did not occur. Conclusions: First-line chemotherapy for HER2-negative patient of mXP showed a promising response rate and an acceptable tolerability profile in the clinical practice in Japanese mGC pts including the elderly. Clinical trial information: UMIN:000006668.

Analysis of KRAS/NRAS, PI3CA, and BRAF mutations in the phase II KSCC0901 study of cetuximab plus S-1 as third-line treatment for metastatic colorectal cancer in Japanese patients

587

Background: Exploratory biomarker studies in the randomized phase III PRIME study suggested that patients with mCRC tumors harboring a mutation (MT) in KRAS or NRAS (beyond KRAS exon 2) do not respond to panitumumab-FOLFOX4 (2013 ASCO, JCO 31, #3511). We reported the efficacy and safety of S-1+. Cetuximab (Cmab) in wild-type (WT)-KRAS (exon 2) mCRC patients (pts) as third-line therapy (KSCC0901) (2012 ASCO, JCO 30, s3558). The full spectrum of characterized RAS (KRAS and NRAS), PI3CA, and BRAFMTs may affect efficacy in S-1+Cmab therapy in Japanese patients. Methods: An analysis was conducted to assess the treatment effect of S-1+Cmab, according to RAS (KRAS and NRAS), PI3CA, and BRAF MTs. The treatment protocol of KSCC0901 was as follows: weekly intravenous Cmab administration of 400 mg/m2 (day 1), 250 mg/m2/week (following day 1) and oral administration of 80 mg/m2/day S-1 on days 1–28 of each 42-day cycle. MTs in KRAS codons 12 and 13 (exon 2) were detected using direct sequencing according to the manufacturer’s instructions. MTs in KRAS codons 61 and 146, NRAS codons 12, 13 and 61, BRAF codon 600 and PIK3CA codons 542, 545, 546, and 1047 were detected using the multiplex PCR-Luminex method-based MEBGEN MT Kit (Medical and Biological Laboratories, Nagoya, Japan). Results: 35 pts were eligible for the analysis. Number of MT cases of KRAS, NRAS, BRAF, and PIK3CA were 4, 2, 2, and 5, respectively. One case had MTs of NRAS and PIK3CA. In 37 cases with WT KRASexon2 pts, median progression-free survival (mPFS) was 5.6 [95% CI: 4.4 – 5.7] months; median overall survival (mOS), 13.5 [8.5-16.5] months; and best overall response (ORR), 37.8%. In the all-WT subgroup 23 cases, mPFS was 5.6 [5.1-7.1] months; mOS, 13.5 [9.3-21.2] months; and best ORR, 47.8%. In 12 patients with any MT, mPFS was 4.0 [2.7-7.5] months; mOS, 5.5[2.9- ] months; and best ORR, 16.7%. Conclusions: These results suggest that also in Japanese patients receiving a Cmab containing regimen, RAS, BRAF and PIK3CA MT status beyond KRAS exon2 might predict non-responders to treatment. Clinical trial information: UMIN000002475.

Phase II study of combination therapy with S-1 and cetuximab in patients with KRAS wild-type unresectable colorectal cancer who had previously received irinotecan, oxaliplatin, and fluoropyrimidines (KSCC0901)

3558

Background: Anti-epidermal growth factor receptor (anti-EGFR) antibodies alone or in combination with irinotecan (Iri) can be considered standard third-line therapy for KRAS wild-type (wKRAS) unresectable colorectal cancer (UNCRC). However, some UNCRC patients (pts) cannot tolerate Iri-containing therapy. S-1, an oral fluorouracil (FU) derivative, enhances the anti-tumor effect by inhibiting dihydropyrimidine dehydrogenase activity and reducing digestive toxicity. Combination therapy with cetuximab (C-mab) may restore 5-FU resistance in 5-FU–resistant CCs. Therefore, we examined the efficacy of S-1+C-mab therapy in wKRAS UNCRC pts, who had previously received Iri, oxaliplatin (OX), and FUs. Methods: The study design was multicenter, single-arm, open-label phase II study. The major inclusion criteria were written informed consent; histologically proven CRC and clinically proven UNCRC; presence of measurable lesions; previous therapy with Iri, OX, and 5-FU; documented progressive disease after 5-FU–based chemotherapy; wKRAS tumors; age ≥ 20 years; performance status (PS) 0–1; and adequate organ function. The treatment protocol was as follows: weekly durable intravenous (DIV) C-mab administration at 400 mg/m2 (day 1) and 250 mg/m2/week (except day 1) and oral administration of 80 mg/m2/day S-1 on days 1–28 of each 42-day cycle. The primary endpoint was progression-free survival (PFS). A sample size of 39 was planned for a threshold PFS of 3.5 months and expected value of 6.0 months, with one-sided alpha of 0.05 and beta of approximately 0.2. Results: One patient was ineligible; 38 pts (PS 0/1, 32/6; 1/2/>3 prior chemotherapy regimens, 4/23/11) were enrolled from 10/2009 to 12/2010. The median PFS (central review) was 5.5 months (90% CI: 4.4 – 5.7); median overall survival (OS), 13.1 months; and the best ORR, 36.8%. The most common grade 3–4 adverse events were neutrophils, hypokalemia, rash, and dry skin. Conclusions: This study showed that S-1+C-mab may be a promising and well-tolerated treatment choice of wKRAS UNCRC, who had previously heavily treated by Iri, OX and FUs.

Adjuvant capecitabine treatment for stage III colon cancer in Japanese patients (KSCC0803)

571

Background: Capecitabine was approved in Japan in 2007 for the adjuvant treatment of stage III colon cancer based on Japanese clinical trial data in advanced and recurrent colorectal and breast cancers as well as data from the Phase III X-ACT trial. For the current study, we aimed to clarify compliance and tolerability of adjuvant treatment with capecitabine in Japanese patients. The study was entered in the UMIN clinical trial registry (UMIN000001444) by the Kyushu Study Group of Clinical Cancer (KSCC). Ethical approval was granted by the institutional review board of each hospital involved.

Methods: Based on completion rates from the X-ACT trial we enrolled 97 patients with R0 stage III colon cancer who had histologically confirmed disease and had undergone curative resection (3D2 lymph node dissection). Patients were given oral capecitabine therapy (2,500 mg/m2/day; days 1–14 q3w; eight cycles) within 8 weeks of surgery. The proportion of patients completing eight cycles of treatment per protocol was the primary endpoint, and adverse event (AE) rate was analyzed as a secondary endpoint.

Results: Treatment completion in the total patient population was 66.0% (64/97 patients; 95% CI: 55.7–75.3%) and in the per protocol population (PPP) was 70.3% (64/91; 95% CI: 59.8–79.5%). AEs leading to treatment discontinuation included hand-foot syndrome (HFS; n=7), hematotoxicity (n=5) and increased hepatic activity (n=4). Grade 3/4 AEs of note included HFS (22.7%), neutropenia (7.2%), diarrhea (2.1%), and increased bilirubin (0.0%). Of note, any treatment delay >3 weeks in the current trial was considered a withdrawal. In the X-ACT trial, delays of any duration were permitted. Using the X-ACT criteria the completion rate for the PPP in this study was 80.2%, comparable to the figure reported in X-ACT.

Conclusions: Our results confirm those of previous global phase III studies and show that capecitabine is well tolerated in both global and Japanese-only populations, with similar high completion rates in both.

[Table: see text]

Analysis of oxaliplatin-induced sensory neurotoxicity (sNT) in patients receiving FIREFOX, alternating regimen of mFOLFOX-6 and FOLFIRI, with metastatic colorectal cancer (MCRC): Results from combined analysis of KSCC0501 and KSCC0701

492

Background: The Kyushu Study Group of Clinical Cancer (KSCC) conducted two phase II trials (KSCC0501 and KSCC0701, Akagi et al. J Clin Oncol. 28:15s, 2010, UMIN ID: 000001342) to evaluate the efficacy and safety of first-line oxaliplatin-based chemotherapy for MCRC. This combined analysis was performed to compare the incidence of oxaliplatin-induced sNT between the two trials.

Methods: Patients (pts) were accrued from 2005 to 2007 in KSCC0501 and from 2007 to 2008 in KSCC0701. Sixty pts received FOLFOX-4 in KSCC0501 and 47 pts received FIREFOX(4 cycles of mFOLFOX-6 alternating with 4 cycles of FOLFIRI) in KSCC0701. All pts were reviewed for efficacy and toxicity (NCI-CTCAEv3.0). Kaplan-Meier analysis was performed to assess the incidence of sNT.

Results: The incidence of sNT was 71.4% with FOLFOX-4 and 36.2% with FIREFOX (Table). The ORR was 34.5% (95% CI F22.5-48.1%) for FOLFOX4 and 58.7% (43.9-73.5%) for FIREFOX. Median PFS was 7.0 M (5.1-9.8 M) with FOLFOX-4 and 10.3 M (7.5-11.9 M) with FIREFOX. MST and 2-year survival were respectively 31.5 M (18.1-40.1 M) and 58.0% for FOLFOX4, versus not determined and 57.1% for FIREFOX. The median no. of treatment cycles was 9 for FOLFOX-4 and 12 for FIREFOX. After 4, 8, and 12 cumulative treatment cycles, the incidence of grade 2+sNT was respectively 24.0%, 30.7%, and 60.5% with FOLFOX-4 versus 6.5%, 6.5%, and 16.0% with FIREFOX.

Conclusions: As first-line oxaliplatin-based chemotherapy for MCRC, FIREFOX caused less oxaliplatin-induced sNT and prolonged the duration of treatment. We have now finished enrollment for study KSCC 0801 (KSCC 0701+bevacizumab) and are following the pts.

[Table: see text]

[Table: see text]