PHASE II STUDIES

A phase II open-label randomized multicenter trialof TSU-68 in combination with S-1 and oxaliplatin versus S-1in combination with oxaliplatin in patients with metastaticcolorectal cancer

Jeeyun Lee & Sang Joon Shin & Ik Joo Chung & Tae Won Kim & Hoo-Geun Chun &

Dong Bok Shin & Yeul Hong Kim & Hong Suk Song & Sae-Won Han & Jong Gwang Kim &

Sun Young Kim & Young Jin Choi & Hyun Cheol Chung

Received: 30 November 2013 /Accepted: 10 January 2014# Springer Science+Business Media New York 2014

Summary BackgroundColorectal cancer (CRC) is the fourthleading cause of cancer-related deaths worldwide. The com-bination of oxaliplatin-based treatments (oxaliplatin plusinfusional 5-fluorouracil and leucovorin [FOLFOX] oroxaliplatin plus capecitabine [CapeOX]) and bevacizumab isa standard chemotherapy regimen for metastatic CRC(mCRC). However, several clinical studies that tested S-1 plus

oxaliplatin (SOX) indicate that SOX is also a treatment optionfor mCRC. TSU-68 is an oral compound that inhibits vascularendothelial growth factor receptor and platelet-derived growthfactor receptor. The recommended dose of TSU-68+SOXwaspreviously determined in a phase I study of mCRC patients.The goal of this trial was to evaluate the efficacy of TSU-68 incombination with SOX. Methods This open-label multicenter

J. LeeDivision of Hematology/Oncology, Department ofMedicine,Samsung Medical Center, Sungkyunkwan UniversitySchool of Medicine, Seoul, South Koreae-mail: jyunlee@skku.edu

S. J. Shin :H. C. ChungDivision of Hematology/Oncology, Department of Medicine,Severance Hospital, Yonsei University College of Medicine, Seoul,South Korea

I. J. ChungDivision of Hematology/Oncology, Department of Medicine,Chonnam National University Hwasun Hospital, Gwangju, SouthKorea

T. W. KimDivision of Hematology/Oncology, Department of Medicine, AsanMedical Center, Seoul, South Korea

H.<G. ChunDivision of Hematology/Oncology, Department of Medicine, SeoulSt. Mary’s Hospital, Banpo, Seoul, South Korea

D. B. ShinDivision of Hematology/Oncology, Department of Medicine,Gachon University Gil Hospital, Incheon, South Korea

Y. H. KimDivision of Hematology/Oncology, Department of Medicine, KoreaUniversity Anam Hospital, Seoul, South Korea

H. S. SongDivision of Hematology/Oncology, Department of Medicine,Keimyung University Dongsan Hospital, Seoul, South Korea

S.<W. HanDivision of Hematology/Oncology, Department of Medicine, SeoulNational University Hospital, Seoul, South Korea

J. G. KimDivision of Hematology/Oncology, Department of Medicine,Kyungpook National University Hospital, Daegu, South Korea

S. Y. KimDivision of Hematology/Oncology, Department of Medicine,National Cancer Center, Goyang-si, South Korea

Y. J. ChoiDivision of Hematology/Oncology, Department of Medicine, PusanNational University Hospital, Pusan, South Korea

H. C. Chung (*)Department of Medical Oncology, Severance Hospital, YonseiUniversity, 250 Seongsanno, Seodaemun-gu, Seoul 120-752, SouthKoreae-mail: unchung8@yuhs.ac

Invest New DrugsDOI 10.1007/s10637-014-0075-8

randomized phase II trial was performed in Korea. Treatment-naive mCRC patients with a performance status of 0 or 1 wererandomized in a 1:1 ratio to receive either TSU-68+SOX orSOX alone. The primary endpoint was progression-free sur-vival (PFS). Results A total of 105 patients (TSU-68+SOX,52 patients; SOX alone, 53 patients) were randomized. Themedian PFS was 7.0 months in the TSU-68+SOX group(hazard ratio [HR], 1.057) and 7.2 months in the SOX group(p=0.8401). The most frequent grade 3 and 4 adverse eventswere thrombocytopenia (9.6 % [TSU-68+SOX] vs. 26.4 %[SOX]), neutropenia (13.5 % [TSU-68+SOX] vs. 15.1 %[SOX]), and anemia (3.8 % [TSU-68+SOX] vs. 13.2 %[SOX]). We observed a difference between the 2 groups forall grades of anemia (15.4 % [TSU-68+SOX] vs. 32.1 %[SOX]), diarrhea (30.8 % [TSU-68+SOX] vs. 47.2 %[SOX]), vomiting (50.0 % [TSU-68+SOX] vs. 26.4 %[SOX]), and chromaturia (23.1 % [TSU-68+SOX] vs. 0.0 %[SOX]). Analysis using a Cox proportional hazard modelshowed that baseline interleukin 6 (IL-6) levels were associ-ated with a survival benefit of TSU-68 (p=0.012). ConclusionTSU-68+SOX had a favorable safety profile. However, TSU-68 did not have a synergistic effect on the efficacy of SOX.The baseline serum IL-6 level could be a prognostic factor forTSU-68 efficacy.

Keywords Colorectal cancer . Chemotherapy . TSU-68

Introduction

Colorectal cancer (CRC) is the fourth leading cause of cancer-related deaths worldwide. A doublet chemotherapy regimenof fluoropyrimidines plus either oxaliplatin or irinotecan, withor without targeted agents, is considered the standard treat-ment regimen for metastatic colorectal cancer (mCRC) pa-tients [1–10]. Bevacizumab is a humanized monoclonal im-munoglobulin (Ig) G1 antibody that targets the circulatingvascular endothelial growth factor (VEGF), and it has beenshown to improve the response rate (RR) and survival inmCRC patients when administered in combination with stan-dard cytotoxic chemotherapy regimens. Large randomizedtrials have shown that first-line treatment with bevacizumabin combination with irinotecan plus bolus 5-fluorouracil ([5-FU], IFL) improves the RR, progression-free survival (PFS),and overall survival (OS), and improves PFS when combinedwith oxaliplatin plus infusional 5-FU and leucovorin ([LV],FOLFOX) or oxaliplatin plus capecitabine (CapeOX) [7].

S-1 is a new-generation oral fluoropyrimidine that is com-posed of tegafur (FT), an oral pro-drug of 5-FU, with 2modulators: gimeracil (5-chloro-2, 4-dihydroxypyridine,[CDHP]), which inhibits degradation of 5-FU from FT bydihydropyrimidine dehydrogenase (DPD), and oteracil potas-sium (Oxo), which inhibits 5-FU phosphorylation in the

digestive tract. S-1 in combination with oxaliplatin (SOX)has been considered a treatment option for mCRC [11, 12].A recent phase III study showed that the efficacy of SOX wasequivalent to that of CapeOX when used as a first-line treat-ment for mCRC patients [13].

TSU-68 ([Z]-5-[(1,2-dihydro-2-oxo-3H-indol-3-ylidene)methyl]-2,4-dimethyl-1H-pyrrole-3-propanoic acid)is a potent anti-angiogenic agent that targets multiple tyrosinekinase receptors including vascular endothelial growth factorreceptor 2 (VEGFR2), platelet-derived growth factor receptorβ (PDGFRβ), and fibroblast growth factor receptor 1(FGFR1) [14]. In a phase II study, TSU-68 showed favorableefficacy and a good safety profile when used to treathepatocellular carcinoma (HCC) after transarterialchemoembolization (TACE) [15]. Based on the findingsfrom in vivo preclinical experiments in xenograft models ofCRC [16], a phase I study was conducted to assess the safetyand to determine the recommended dose of TSU-68 incombination with SOX [17]. Based on the results of thisstudy, the recommended regimen was 35 mg/m2 S-1 twice aday on days 1 to 14, followed by a 7-day recovery period,and 130 mg/m2 oxaliplatin on day 1 with 200 mg TSU-68twice daily.

This randomized phase II study was conducted tocompare the efficacy and tolerability of SOX treatmentwith or without TSU-68 as first-line chemotherapy formCRC patients.

Methods

Study design and participants

This was an open-label, randomized, controlled, multicenterphase II study that compared the efficacy of the combinationof TSU-68 and SOX with that of SOX alone in treatment-naïve mCRC patients. The patients were randomized andtreated at 12 centers in Korea. Patients with mCRC wereconsidered eligible if they met the following criteria: a histo-logical examination indicating adenocarcinoma; measurableor assessable lesions; an Eastern Cooperative OncologyGroup (ECOG) performance status (PS) of 0–1; aged at least18 years old; no prior chemotherapy treatment, excludingchemotherapies in the adjuvant or neo-adjuvant setting fornon-metastatic disease (M0) where the patient relapsed at6 months or more (> 180 days) after completion of thechemotherapy; adequate hematological, hepatic, and renalfunctions. All patients provided written informed consentbefore entering the study, and institutional review boards ofall participating institutions approved the study protocol.

After the patients provided written informed consent, and itwas confirmed that the patients met all of the inclusion andnone of the exclusion criteria, each patient was centrally

Invest New Drugs

randomized via the interactive web response system (IWRS)based on a stochastic minimization technique to receive eitherTSU-68+SOX or SOX alone. Treatment was initiated within7 days after randomization.

This study was carried out in accordance with Good Clin-ical Practice (GCP) and the Declaration of Helsinki guide-lines. Documented approval from appropriate ethics commit-tees and institutional review boards was obtained for all par-ticipating centers.

Study treatment

Patients who met the inclusion criteria and did not meet theexclusion criteria were randomized in a 1:1 ratio to receiveone of the following treatment regimens: 200 mg TSU-68administered orally twice a day (within approximately 1 hafter a meal) in the morning and evening on days 1 to 21;35 mg/m2 S-1 administered orally twice a day (within approx-imately 1 h after a meal) in the morning and evening on days 1to 14, followed by a 7-day recovery period, and 130 mg/m2

oxaliplatin delivered intravenously via a 2 h infusion on day 1,either with or without 200 mg TSU-68 administered orallytwice a day (within approximately 1 h after a meal) in themorning and evening on days 1 to 21.. The regimen wasrepeated every 3 weeks. Study treatment continued until oneof the following occurred: documented progressive disease(PD), an unacceptable adverse event (AE), or withdrawal ofconsent. Patients who ended treatment without PD were con-tinuously monitored for tumor response every 6 weeks fromthe time they stopped receiving the study treatment until PDwas found. After the study treatment ended, the survival statusof patients was monitored every 12 weeks, starting at the timeof treatment discontinuation and continuing until their deathor for up to 2 years after the day on which the last patient wasrandomized, whichever occurred first. Doses were reducedaccording to the protocol.

Objective

The primary objective was to evaluate the PFS for bothtreatment groups. The secondary objectives were to compareRR, OS, disease control rate (DCR), and safety. Biomarkeranalyses were performed to determine the relationship be-tween treatment efficacy and the levels of angiogenesis factors(serum levels of VEGF, PDGF-AA, PDGF-BB, and vascularcell adhesion molecule 1 [VCAM-1]; plasma levels of VEGF,PDGF-AA, and PDGF-BB; and serum lactate dehydrogenase[LDH] isozyme levels) and inflammatory markers (IL-6, IL-8,and C-reactive protein [CRP]). Baseline measurements ofthese angiogenesis factors and inflammatory markers weretaken within 28 days before the study treatment started, onday 1 of cycle 1, between day 15 of cycle 1 (±3 days) and day1 of the next cycle, and at the end of the study treatment.

Statistical considerations

PFS was defined as the time from the date of randomization tothe date of disease progression or death from any cause thatoccurred during the study, whichever happened first. If con-firmation of disease progression was not available or if thepatient died, the PFS was censored on the last date of tumorassessment (the last date on which images were obtained) orthe date of death, whichever was first. If tumor assessmentdata was absent after randomization, the PFS was censored onday 1 (date of randomization). The statistical test for analysisof PFS involved an unstratified log-rank test with a 5 % one-sided significance level. The hazard ratio (HR) for the treat-ment arm was estimated with a 95 % confidence interval (CI)by using a multivariate Cox proportional hazard model thatincluded stratification factors.

From the results of phase II studies of the SOX regimen,the median PFS with SOX in the study population was esti-mated to be approximately 7.0 months. Themedian PFS of thepopulation that received TSU-68 in combination with SOXwas projected to be 13.0 months. The resulting HR (TSU-68in combination with SOX vs. SOX alone) was 0.538, which isequivalent to an improvement in PFS of approximately 85 %.The study had 80% power to test the superiority of TSU-68 incombination with SOX versus SOX alone using a log-ranktest with a 1-sided probability of 5%, when the HRwas 0.538,and the targeted number of PFS events was 70. To achieve thistarget number of events, we required a total of 94 randomizedpatients who were given the indicated treatments, assuming anaccrual of 1.2 years and a minimum follow-up period of1.0 year. With the added consideration that up to 5 % ofpatients may be excluded from the population (randomizedbut not treated), a total of 100 patients were required to meetthe primary objective of this study.

The population studied for the evaluation of treatmentefficacy was the “As-Treated Population,” (ATP) which in-cluded all randomized patients without major protocol devia-tions who were treated with at least 1 dose of the studytreatment in each treatment arm, even if the treatment armwas different from the one assigned during randomization.Safety analysis was performed in the “Safety Population,”which included all randomized patients who were treated withat least one dose of the study treatment in each treatment arm.

Results

A total of 105 patients were randomized to the treatmentgroups (TSU-68+SOX arm, 52 patients; SOX alone arm, 53patients) between March 2010 and December 2010 (Fig. 1).All patients randomized to the study were included in theSafety Population. One patient from the Safety Populationwho received TSU-68+SOX failed to meet the inclusion

Invest New Drugs

criteria (because of prior chemotherapy treatment, excludingchemotherapy in the adjuvant or neo-adjuvant setting for non-metastatic disease) and was excluded from the ATP. Baselinepatient characteristics are provided in Table 1. No significantdifferences in baseline clinicopathologic features were ob-served between the 2 arms. Most patients (~80 %) had morethan 1 metastatic lesion at the time of study entry.

Treatment outcome

Complete response (CR) was seen in 2 patients in the TSU-68+SOX group. A total of 18 patients showed a CR or partialresponse (PR) in the TSU-68+SOX group, resulting in anORR of 35.3 %, compared to 29 patients in the SOX group,resulting in an ORR of 54.7 % (Fisher’s exact test: p=0.036).Kaplan-Meier plots for PFS are shown in Fig. 2. The medianPFS was 7.0 months for the TSU-68+SOX arm and7.2 months for the SOX arm (HR=1.057, 95 % CI=0.619–1.806; p=0.840) as assessed by an independent review com-mittee. The independent review committee also used the PFSdata to perform pre-planned subgroup analyses (Fig. 3). Fac-tors such as gender, age, primary site location, number ofmetastatic sites, and best tumor response revealed no notice-able benefits of the TSU-68+SOX treatment, as determinedby Cox regression analysis. The median OS in the TSU-68+SOX arm was 16.7 months vs. 16.4 months in the SOX arm(HR=1.051, 95 % CI=0.622–1.778; p=0.852, figure notshown).

Toxicity profile

All patients experienced adverse events during the study, with38 patients (18 patients who received TSU-68+SOX vs. 18patients who received SOX alone) experiencing serious ad-verse events (SAEs). The major adverse events are listed inTable 2. The most frequent grade 3 and 4 events were throm-bocytopenia (9.6 % [TSU-68+SOX] vs. 26.4 % [SOX]),neutropenia (13.5 % [TSU-68+SOX] vs. 15.1 % [SOX]),and anemia (3.8 % [TSU-68+SOX] vs. 13.2 % [SOX]). We

observed differences between the 2 groups in all grades ofanemia (15.4 % [TSU-68+SOX] vs. 32.1 % [SOX]), diarrhea(30.8 % [TSU-68+SOX] vs. 47.2 % [SOX]), vomiting(50.0 % [TSU-68+SOX] vs. 26.4 % [SOX]), and chromaturia(23.1 % [TSU-68+SOX] vs. 0.0 % [SOX]).

Reason for treatment discontinuation

Reasons for treatment discontinuation are shown in Table 3.Most patients discontinued both treatments because clinical

Fig. 1 CONSORT diagram. Onepatient in the TSU-68+SOXgroup was excluded from the As-Treated Population because ofmajor deviation from the protocol

Table 1 Patient characteristics

TSU-68 plusSOX (N=52)

SOX(N=53)

Characteristic n (%) n (%)

Gender Male 32 (61.5) 39 (73.6)

Female 20 (38.5) 14 (26.4)

Age Mean 58.1 58.1

Median 58.0 60.0

Range 31.0–79.0 28.0–74.0

BSA Mean 1.7 1.7

Median 1.65 1.72

Range 1.25–1.88 1.25–2.01

ECOG PS 0 23 (44.2) 28 (52.8)

1 29 (55.8) 25 (47.2)

Primary tumor site Colon 33 (63.5) 28 (52.8)

Rectum 14 (26.9) 15 (28.3)

Rectosigmoid colon 5 (9.6) 10 (18.9)

Tissue Type Well- differentiated 20 (40.0) 23 (45.1)

Moderately-differentiated

13 (26.0) 15 (29.4)

Poorly- Differentiated 10 (20.0) 7 (13.7)

Mucinous carcinoma 3 (6.0) 3 (5.9)

Signet- ring cellcarcinoma

4 (8.0) 3 (5.9)

Other

No. of metastaticsite

1 11(21.2) 11 (20.8)

> 1 41 (78.8) 42 (79.2)

Invest New Drugs

imaging results revealed disease progression. The percentageof patients who discontinued treatment because of adverseevents was less than 15 % of the total patients in each group.The study was discontinued in March 2012, and the fewpatients who continued the study treatment without experienc-ing disease progression were transitioned to standard treat-ment, with informed consent.

Biomarker analyses

To determine the predictive values of various biomarkers, therelationships between treatment efficacy and baseline bio-marker levels were analyzed. Table 4 shows the results ofanalysis using a Cox model that included the treatment group,

number of metastases, and baseline value of the indicatedbiological marker (Table 4). Assessment of PFS by the inde-pendent review committee revealed statistically significantbaseline biological marker effects of serum PDGF-BB levels(p=0.033), serum VCAM-1 levels (p=0.011), plasma VEGF

Fig. 2 Kaplan-Meier analysis. Progression-free survival (PFS) for 104patients in the study. An independent review committee examined theimages. There were 51 patients in the TSU-68+SOX group and 53patients in the SOX alone group

Fig. 3 Progression-free survivalof selected subgroups accordingto background. A forest plotdepicting the hazard ratio for PFS(TSU-68+SOX group over SOXalone group) for each subgroup

Table 2 Incidence of adverse events

Adverse event TSU-68+SOX (N=52) SOX (N=53)

Grade (%) Grade (%)

Any ≥ 3 Any ≥ 3

Anemia 15.4 3.8 32.1 13.2

Neutropenia 28.8 13.5 41.5 15.1

Thrombocytopenia 63.5 9.6 69.8 26.4

Abdominal pain 36.5 1.9 41.5 9.4

Constipation 30.8 1.9 35.8 1.9

Diarrhea 30.8 0 47.2 11.3

Dyspepsia 30.8 0 18.9 0

Nausea 67.3 3.8 66.0 5.7

Stomatitis 22.6 0 21.2 1.9

Vomiting 50.0 0 26.4 1.9

Fatigue 30.8 3.8 45.3 7.5

Pain 25.0 1.9 13.2 0

Peripheral sensory neuropathy 59.6 3.8 69.8 3.8

Chromaturia 23.1 0 0 0

Skin hyper pigmentation 42.3 0 26.4 0

Listed are adverse events, as defined by National Cancer Institute Com-mon Terminology Criteria (version 3.0), that occurred in at least 20 % ofpatients in the both groups

Invest New Drugs

levels (p=0.013), and inflammatory factor IL-8 levels (p=0.009). There were no statistically significant alterations ofbiological markers by the treatment that indicated an effect ofspecific levels of each marker on PFS. For OS, statisticallysignificant effects of baseline biological markers were ob-served for serum VEGF (p=0.011), plasma VEGF (p=0.001), serum VCAM-1 (p<0.001), serum LDH isozymes(p=0.003), and the inflammatory markers IL-6 (p=0.048),IL-8 (p<0.001), and CRP (p=0.015). There was a statisticallysignificant interaction between the treatment and baseline IL-6levels (p=0.012), indicating that the effect on OS differed as afunction of IL-6 levels.

Discussion

This is the first study to evaluate whether the experimentaltreatment combination of TSU-68 and SOX has a synergisticeffect on efficacy in mCRC patients. We randomized 105patients into groups to receive either SOX alone or TSU-68+SOX as first-line chemotherapy for mCRC.

Addition of TSU-68 did not result in additional toxicity inthis study (Table 3). The frequency of grade 3 and 4 throm-bocytopenia, which is one of the dose-limiting toxicities ofoxaliplatin, did not increase with the addition of TSU-68.Interestingly, the incidence of grade 3/4 thrombocytopeniawas lower (9.6 %) in the TSU-68+SOX arm compared tothe SOX arm (26.4 %). The frequency of grade 3 and 4thrombocytopenia observed in our patient cohort was similarto that reported in a previous SOX/CapeOX phase III trial(22 %) [13]. Moreover, the incidence of peripheral sensoryneuropathy of any grade was similar between the TSU-68+SOX arm and the SOX arm (59.6 % [TSU-68+SOX] vs.69.8 % [SOX]). No treatment-related mortalities were ob-served. Hence, the addition of TSU-68 to SOX did not in-crease toxicity. Regarding the reasons for treatment discontin-uation, most patients in both groups were able to continue thestudy treatment until presented with radiologic evidence ofdisease progression. Therefore, TSU-68+SOX can be consid-ered as a highly tolerable regimen for mCRC patients, similarto treatment with SOX alone.

No difference was observed in the median PFS, the primaryendpoint of this study, between the 2 arms (median PFS,7.0 months [TSU-68+SOX] vs. 7.2 months [SOX], HR=1.057, 95 % CI=0.619–1.806, p=0.840]; Fig. 2). Factors such

Table 3 Reason for treatment discontinuation

Reason for treatment discontinuation TSU-68+SOX(N=51)

SOX(N=53)

N % N %

Continuation1 1 2.0 2 3.8

Discontinuation 50 98.0 51 96.2

PD-Radiologic Progression 33 64.7 35 66.0

PD-Clinical Progression 2 3.9 1 1.9

Adverse Events 7 13.7 2 3.8

Patients withdrew consent 3 5.9 7 13.2

Investigator Discretion 4 7.8 5 9.4

Lost to Follow-up 1 2.0 0 0

Low-compliance 0 0 1 1.9

1 The study was discontinued in March 2012. The few patients whocontinued the study treatment without disease progression made thetransition to standard treatment with informed consent.

Table 4 Progression-free survivaland overall survival: biologiclmarker effect

1 Analysis using a using a Coxmodel including treatment group,number of metastases, and base-line value of biological marker2 Analysis using a Cox model in-cluding treatment group, numberof metastases, baseline value ofbiological marker and treatmentgroup x baseline biologicalmarker interaction

PFS OS

Hazardratio

P-Value1 InteractionP-value2

Hazardratio

P-value InteractionP-value

Serum

VEGF 1.00008 0.817 0.191 1.00069 0.011 0.953

PDGF-AA 1.00008 0.502 0.192 1.00010 0.350 0.451

PDGF-BB 1.00023 0.033 0.997 1.00015 0.171 0.922

VCAM-1 1.00109 0.011 0.606 1.00165 0.000 0.762

LDH isozymes 1.00011 0.092 0.755 1.00021 0.003 0.647

Plasma

VEGF 1.00378 0.013 0.349 1.00303 0.001 0.975

PDGF-AA 1.00042 0.111 0.724 1.00016 0.502 0.686

PDGF-BB 1.00026 0.538 0.201 0.99976 0.563 0.168

Inflammatory marker

IL-6 1.00499 0.094 0.123 1.00527 0.048 0.012

IL-8 1.00497 0.009 0.531 1.00717 0.000 0.409

CRP 1.00343 0.378 0.564 1.00867 0.015 0.351

Invest New Drugs

as gender, age, primary site location, number of metastaticsites, and best tumor response showed no prominent benefitsof the TSU-68+SOX treatment in the Cox regression analysis.No significant difference in survival was observed betweenthe 2 arms (median OS, 16.7 months [TSU-68+SOX] vs.16.4 months [SOX], HR=1.051, 95 % CI=0.622–1.778;p=0.852).

Recently, a randomized phase III trial was performed tocompare the efficacy of SOX vs. CapeOX as a first-linetreatment for mCRC. The trial showed that the SOX regimenwas superior to CapeOX in terms of PFS (8.5 months for SOXvs. 6.7 months for CapeOX) [13]. The median PFS obtainedin this study for both TSU-68+SOX (7.0 months) and SOX(7.2 months) were similar to the CapeOX arm in the previousphase III study, but were shorter than the SOX arm. Subsetanalysis of the NO16966 study to compare PFS in patientsreceiving CapeOX/FOLFOX alone vs. CapeOX/FOLFOXplus bevacizumab showed that the median PFS was9.4 months for patients receiving FOLFOX or CapeOX withbevacizumab and 8.0 months for the placebo group [6].Therefore, the PFS obtained in this randomized phase II studyfor both TSU-68+SOX and SOX alone were lower than thosereported in other large phase III trials.

To identify a subgroup of patients who may potentiallybenefit from the addition of TSU-68 to standard chemothera-py, we performed pre-planned analyses of serial VEGF,PDGF-AA, and IL-6 measurements to test them as possiblesurrogate markers of treatment efficacy. Baseline measure-ments were taken within 28 days before the start of the studytreatment, on day 1 of the cycle 1, between day 15 of the cycle1 (±3 days) and day 1 of the next cycle, and at the end of thestudy treatment. Vatalanib is a tyrosine kinase inhibitor similarto TSU-68 that targets VEGFR-1, VEGFR-3, and VEGFs. Ina phase I study of vatalanib, the serum levels of angiogenesisbiomarkers and activated endothelial cells were evaluated toassess biological activity of the drug [18]. A dose-dependentincrease in VEGF-A was observed in 63 mCRC patientstreated with vatalanib [18]. In our study, the baseline IL-6level was associated with a survival benefit of TSU-68 (p=0.012) in mCRC patients (Table 4). However, the other bio-markers did not show any correlation with treatment outcome.Additionally, many controversial studies have reported theefficacy of soluble biomarkers as predictive surrogate markersfor treatment response to tyrosine kinase inhibitors againstangiogenesis. A large panel of 174 cytokines were analyzedin renal cell carcinoma (RCC) patients before and after treat-ment with sunitinib, and only baseline tumor necrosis factorα(TNF-α) and matrix metalloproteinase 9 (MMP-9) levelswere found to be predictive of sunitinib efficacy [19]. How-ever, no correlations were observed between the levels ofsoluble cytokines and response to sunitinib in patients withRCC [20]. Therefore, it is difficult to conclude that mCRCpatients with high baseline IL-6 levels may benefit from the

addition of TSU-68 to standard oxaliplatin-basedchemotherapy.

The ORR was higher for patients who received SOX thanfor those who received TSU-68+SOX (TSU-68+SOX,35.3% vs. SOX, 54.7%). Furthermore, with regard to adverseevents, there was a higher incidence of vomiting, skin hyper-pigmentation, dyspepsia, pain, facial edema, and chromaturiain patients who received TSU-68+SOX. Conversely, therewas a high incidence of diarrhea, fatigue, decreased neutrophilcount, anemia, pyrexia, and increased blood bilirubin in pa-tients who received SOX alone. Diarrhea, anemia, and neu-tropenia are S-1-specific adverse drug reactions, and they hada low incidence in the TSU-68+SOX group. When the effectof TSU-68 on both the ORR and adverse events is considered,it can be said that TSU-68 decreases the efficacy of SOXtreatment. TSU-68 is thought to induce CYP1A1 and 1A2[21], while FT, the main compound of S-1, is mainly metab-olized by CYP2A6 [22]. Since CYP1A1 and CYP1A2 canmetabolize FT, a potential drug interaction between TSU-68and S-1 may influence the efficacy of the TSU-68+SOXcombination. However, we propose that further analysis isnecessary, including an examination of the effect of patientbackground and differential pharmacokinetics on the efficacyof the combination.

In conclusion, TSU-68+SOX did not result in a significantprolongation of PFS compared to SOX alone. However, TSU-68+SOX treatment was shown to have a favorable safetyprofile.

Clinical trials This study is registered with JAPIC Clinical Trial Infor-mation, number JapicCTI-111403.

Conflict of interest statement The authors declare no conflict ofinterest.

References

1. De Gramont A, Figer A, Seymour M, Homerin M, Hmissi A,Cassidy J, Boni C, Cortes-Funes H, Cervantes A, Freyer G,Papamichael D, Le Bail N, Louvet C, Hendler D, de Braud F,Wilson C, Morvan F, Bonetti A (2000) Leucovorin and fluorouracilwith or without oxaliplatin as first-line treatment in advanced colo-rectal cancer. J Clin Oncol 18:2938–2947

2. Douillard JY, Cunningham D, Roth AD, Navarro M, James RD,Karasek P, Jandik P, Iveson T, Carmichael J, Alakl M, Gruia G,Awad L, Rougier P (2000) Irinotecan combined with fluorouracilcompared with fluorouracil alone as first-line treatment for metastaticcolorectal cancer: a multicentre randomised trial. Lancet 355:1041–1047

3. Giantonio BJ, Catalano PJ, Meropol NJ, O’Dwyer PJ, Mitchell EP,Alberts SR, Schwartz MA, Benson AB 3rd (2007) Bevacizumab incombination with oxaliplatin, fluorouracil, and leucovorin(FOLFOX4) for previously treated metastatic colorectal cancer: re-sults from the eastern cooperative oncology group study E3200. JClin Oncol 25:1539–1544

Invest New Drugs

4. Grothey A, SugrueMM, Purdie DM, DongW, Sargent D, Hedrick E,Kozloff M (2008) Bevacizumab beyond first progression is associ-ated with prolonged overall survival in metastatic colorectal cancer:results from a large observational cohort study (BRiTE). J Clin Oncol26:5326–5334

5. Hochster HS, Hart LL, Ramanathan RK, Childs BH, Hainsworth JD,Cohn AL, Wong L, Fehrenbacher L, Abubakr Y, Saif MW,Schwartzberg L, Hedrick E (2008) Safety and efficacy of oxaliplatinand fluoropyrimidine regimens with or without bevacizumab as first-line treatment of metastatic colorectal cancer: results of the TREEstudy. J Clin Oncol 26:3523–3529

6. Saltz LB, Clarke S, Diaz-Rubio E, Scheithauer W, Figer A, Wong R,Koski S, Lichinitser M, Yang TS, Rivera F, Couture F, Sirzen F,Cassidy J (2008) Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectalcancer: a randomized phase III study. J Clin Oncol 26:2013–2019

7. Hurwitz H, Fehrenbacher L, NovotnyW, Cartwright T, Hainsworth J,HeimW, Berlin J, Baron A, Griffing S, Holmgren E, Ferrara N, FyfeG, Rogers B, Ross R, Kabbinavar F (2004) Bevacizumab plusirinotecan, fluorouracil, and leucovorin for metastatic colorectal can-cer. N Engl J Med 350:2335–2342

8. Tournigand C, Andre T, Achille E, Lledo G, Flesh M, Mery-MignardD, Quinaux E, Couteau C, Buyse M, Ganem G, Landi B, Colin P,Louvet C, de Gramont A (2004) FOLFIRI followed by FOLFOX6 orthe reverse sequence in advanced colorectal cancer: a randomizedGERCOR study. J Clin Oncol 22:229–237

9. Van Cutsem E, Rivera F, Berry S, Kretzschmar A, Michael M,DiBartolomeo M, Mazier MA, Canon JL, Georgoulias V, PeetersM, Bridgewater J, CunninghamD (2009) Safety and efficacy of first-line bevacizumab with FOLFOX, XELOX, FOLFIRI andfluoropyrimidines in metastatic colorectal cancer: the BEAT study.Ann Oncol 20:1842–1847

10. Van Cutsem E, Kohne CH, Lang I, Folprecht G, Nowacki MP,Cascinu S, Shchepotin I, Maurel J, Cunningham D, Tejpar S,Schlichting M, Zubel A, Celik I, Rougier P, Ciardiello F (2011)Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-linetreatment for metastatic colorectal cancer: updated analysis of overallsurvival according to tumor KRAS and BRAFmutation status. J ClinOncol 29:2011–2019

11. Yamada Y, Tahara M, Miya T, Satoh T, Shirao K, Shimada Y, OhtsuA, Sasaki Y, Tanigawara Y (2008) Phase I/II study of oxaliplatin withoral S-1 as first-line therapy for patients with metastatic colorectalcancer. Br J Cancer 98:1034–1038

12. Chung KY, Saito K, Zergebel C, Hollywood E, Segal M, Saltz LB(2011) Phase I study of two schedules of oral S-1 in combinationwithfixed doses of oxaliplatin and bevacizumab in patients with advancedsolid tumors. Oncology 81:65–72

13. HongYS, Park YS, LimHY, Lee J, Kim TW, KimKP, Kim SY, BaekJY, Kim JH, Lee KW, Chung IJ, Cho SH, Lee KH, Shin SJ, Kang HJ,Shin DB, Jo SJ, Lee JW (2012) S-1 plus oxaliplatin versus

capecitabine plus oxaliplatin for first-line treatment of patients withmetastatic colorectal cancer: a randomised, non-inferiority phase 3trial. Lancet Oncol 13:1125–1132

14. Laird AD, Vajkoczy P, Shawver LK, Thurnher A, Liang C,Mohammadi M, Schlessinger J, Ullrich A, Hubbard SR, Blake RA,Fong TA, Strawn LM, Sun L, Tang C, Hawtin R, Tang F, Shenoy N,Hirth KP, McMahon G, Cherrington (2000) SU6668 is a potentantiangiogenic and antitumor agent that induces regression ofestablished tumors. Cancer Res 60:4152–4160

15. Inaba Y, Kanai F, Aramaki T, Yamamoto T, Tanaka T, Yamakado K,Kaneko S, Kudo M, Imanaka K, Kora S, Nishida N, Kawai N, SekiH, Matsui O, Arioka H, Arai Y (2013) A randomised phase II studyof TSU-68 in patients with hepatocellular carcinoma treated bytransarterial chemoembolisation. Eur J Cancer 49:2832–2840

16. YamamotoM, Kikuchi H, OhtaM, Kawabata T, Hiramatsu Y, KondoK, Baba M, Kamiya K, Tanaka T, Kitagawa M, Konno H (2008)TSU68 prevents liver metastasis of colon cancer xenografts by mod-ulating the premetastatic niche. Cancer Res 68:9754–9762

17. Shin SJ, Jung M, Jeung HC, Kim HR, Rha SY, Roh JK, Chung HC,Ahn JB (2012) A phase I pharmacokinetic study of TSU-68 (amultiple tyrosine kinase inhibitor of VEGFR-2, FGF and PDFG) incombination with S-1 and oxaliplatin in metastatic colorectal cancerpatients previously treated with chemotherapy. Invest NewDrugs 30:1501–1510

18. Drevs J, Zirrgiebel U, Schmidt-Gersbach CI, Mross K, Medinger M,Lee L, Pinheiro J, Wood J, Thomas AL, Unger C, Henry A, StewardWP, Laurent D, Lebwohl D, Dugan M, Marme D (2005) Solublemarkers for the assessment of biological activity with PTK787/ZK222584 (PTK/ZK), a vascular endothelial growth factor receptor(VEGFR) tyrosine kinase inhibitor in patients with advanced colo-rectal cancer from two phase I trials. Ann Oncol 16:558–565

19. Perez-Gracia JL, Prior C, Guillen-Grima F, Segura V, Gonzalez A,Panizo A, Melero I, Grande-Pulido E, Gurpide A, Gil-Bazo I, CalvoA (2009) Identification of TNF-alpha and MMP-9 as potential base-line predictive serum markers of sunitinib activity in patients withrenal cell carcinoma using a human cytokine array. Br J Cancer 101:1876–1883

20. Farace F, Gross-Goupil M, Tournay E, TaylorM, Vimond N, JacquesN, Billiot F, Mauguen A, Hill C, Escudier B (2011) Levels ofcirculating CD45(dim)CD34(+)VEGFR2(+) progenitor cells corre-late with outcome in metastatic renal cell carcinoma patients treatedwith tyrosine kinase inhibitors. Br J Cancer 104:1144–1150

21. Kitamura R, Matsuoka K, Nagayama S, Otagiri M (2008) Time-dependent induction of rat hepatic CYP1A1 and CYP1A2 expressionafter single-dose administration of the anti-angiogenic agent TSU-68.Drug Metab Pharmacokinet 23:421–427

22. Ikeda K, Yoshisue K, Matsushima E, Nagayama S, Kobayashi K,Tyson CA, Chiba K, Kawaguchi Y (2000) Bioactivation of tegafur to5-fluorouracil is catalyzed by cytochrome P-450 2A6 in human livermicrosomes in vitro. Clin Cancer Res 6:4409–4415

Invest New Drugs