Bevacizumab Combined with S‐1 and Raltitrexed for Patients with Metastatic Colorectal Cancer Refractory to Standard Therapies: A Phase II Study

Abstract

Lessons Learned

• Bevacizumab combined with S‐1 and raltitrexed demonstrated positive antitumor efficacy and acceptable toxicity.

• This combination might represent a treatment option for refractory metastatic colorectal cancer.

Background

In patients with metastatic colorectal cancer (mCRC) refractory to standard therapies, S‐1 plus raltitrexed showed a good objective response rate (ORR) and significant survival benefit in our previous study. In the present study, we assessed the activity and safety of bevacizumab combined with S‐1 and raltitrexed.

Methods

This investigator‐initiated, open‐label, single‐arm, phase II trial was performed at West China Hospital in China. Patients with mCRC who had disease progression after fluoropyrimidine, irinotecan, and oxaliplatin and had at least one measurable lesion were eligible for this trial. Anti–epidermal growth factor receptor (EGFR) (for tumors with wild‐type RAS) and anti–vascular endothelial growth factor (VEGF) therapy in the first or second line was allowed, but patients who had been treated with bevacizumab across two consecutive chemotherapy regimens were excluded. Patients received bevacizumab (7.5 mg/kg on day 1), oral S‐1 (80–120 mg per day for 14 days), and raltitrexed (3 mg/m² on day 1) every 3 weeks. The primary endpoint was ORR. Secondary endpoints included progression‐free survival (PFS), overall survival (OS), and toxicity.

Results

From September 2015 to November 2019, 44 patients were enrolled. Tumor response evaluation was available in 44 patients at the time of the analysis. There were no complete responses; the ORR was 15.9%, and the disease control rate was 54.5%. Median PFS and OS were 110 days (95% confidence interval [CI], 65.0–155.0) and 367 days (95% CI, 310.4–423.6), respectively. The combination was well tolerated.

Conclusion

Bevacizumab combined with S‐1 and raltitrexed showed promising antitumor activity and safety in refractory mCRC.

Discussion

The incidence and mortality of colorectal cancer (CRC) in China are on the rise year by year [1]. Once the disease develops into mCRC, the 5‐year survival rate is less than 15%. Although the multidisciplinary approach of surgery, radiotherapy, chemotherapy, and targeted therapy significantly prolongs survival, OS remains less than 3 years [2].

The initial treatment of mCRC often includes combination chemotherapy, consisting of fluorouracil (5‐FU), oxaliplatin, and irinotecan plus a biologic drug (an EGFR inhibitor or a VEGF inhibitor). If disease progression occurs after the initial treatment, newer agents such as regorafenib, fruquintinib, and TAS‐102 are now available. However, the clinical efficacy of these agents is still unsatisfactory. Respectively, these agents only improve median PFS by 0.2–1.9 months and median OS by 0.7–2.7 months compared with best supportive care [3]. More effective regimens are still needed to improve survival in refractory mCRC.

Dihydropyrimidine dehydrogenase (DPD) is the rate‐limiting enzyme of 5‐FU catabolic pathway. Thymidylate synthase (TS) is the target of 5‐FU's antitumor mechanism. Several studies have shown the upregulation of the two enzymes after the use of 5‐FU in CRC, which may be related to the onset of 5‐FU resistance. Preliminary research in our center reported a 13% ORR and promising survival benefit combining S‐1 (containing a DPD inhibitor) plus raltitrexed (a TS inhibitor) in refractory mCRC [4]. In that study, median PFS and median OS were 107 days (95% CI, 96.3–117.7) and 373 days (95% CI, 226.2–519.8), respectively. Moreover, sustained anti‐VEGF/VEGF receptor therapy has become an important management strategy for mCRC in recent years [5]. However, little is known about the effects of continuous use of angiogenesis inhibitors combined with chemotherapy in third‐ or later‐line therapy.

In this study, bevacizumab combined with S‐1 and raltitrexed demonstrated manageable toxicity and promising antitumor activity in refractory mCRC. Our study showed a relatively higher ORR of 15.9% (7/44) compared with regorafenib, fruquintinib, and TAS‐102 (all less than 5%), similar median PFS (3.7 months vs. 1.9–3.7 months), and longer median OS (12.2 months vs. 5.3–9.3 months) (Fig. 1). Moreover, in 15 patients who had never used bevacizumab before, four patients (26.7%) experienced partial response. Larger sample size and randomized controlled study design are needed to evaluated the true efficacy of this regimen.

Figure 1.

Survival graphs. (A): Progression‐free survival. (B): Overall survival. Abbreviations: CI, confidence interval; mOS, median overall survival; mPFS, median progression‐free survival.

Trial Information

Disease	Colorectal cancer
Stage of Disease/Treatment	Metastatic/advanced
Prior Therapy	More than two prior regimens
Type of Study	Phase II, single arm
Primary Endpoint	Overall response rate
Secondary Endpoints	Progression‐free survival, overall survival, toxicity
Additional Details of Endpoints or Study Design
Study design and participants : This investigator‐initiated, single‐arm, prospective phase II trial was performed in the Cancer Center, West China Hospital, Sichuan University, China. Eligible patients were over 18 years of age; had pathologically confirmed unresectable and metastatic colorectal adenocarcinoma with at least one measurable lesion according to RECIST version 1.1 criteria; had Eastern Cooperative Oncology Group (ECOG) performance status of 0–2 and expected survival more than 12 weeks; and were refractory or intolerant to 5‐fluorouracil (or capecitabine), oxaliplatin, and irinotecan. Anti‐EGFR therapy (for tumors with wild‐type RAS) and anti‐VEGF in the first or second line were allowed, but patients who have been treated with bevacizumab across two consecutive chemotherapy regimens were to be excluded. Additionally, patient should have adequate bone marrow, liver, and renal function. Exclusion criteria were evidence of brain metastases, serious uncontrolled internal disease or acute infection, history of chronic diarrhea or complete intestinal obstruction, extensive surgery within 28 days before enrolment, bleeding tendency, and known allergy to bevacizumab or S‐1 or raltitrexed. All patients were informed of the investigational nature of this study and provided written informed consent before inclusion.
Procedures : Patients received bevacizumab (7.5 mg/kg on day 1), oral S‐1 (80–120 mg per day for 14 days) and raltitrexed (3 mg/m2 on day 1) every 3 weeks. The S‐1 dose was calculated according to body surface area (BSA) as follows: BSA <1.25 m2, 80 mg/day; BSA ≥1.25 m2 but <1.5 m2, 100 mg/day; and BSA ≥1.5 m2, 120 mg/day. The protocol recommended treatment until documented disease progression (PD), unacceptable toxicity, or unwillingness to continue treatment. Physical examination, vital signs, and clinical laboratory tests were done every 3 weeks until progression. Response was determined by the investigator every 6 weeks according to RECIST version 1.1. Adverse events were assessed by the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. Survival follow‐up was carried out every 12 weeks from the time when patients stopped treatment until death or trial cutoff date for data collection.
Endpoints and Statistical Analysis : The primary endpoint was ORR. Secondary endpoints were PFS, OS, and toxicity. Based on Simon's two‐stage designs with p 0 = 5%, p 1 = 20%, α = .05, and β = .10, 21 patients were planned to be enrolled in phase I. If one of these patients achieved complete response (CR) or partial response (PR), another 20 patients (in total over 41 patients) were to be included in phase II. If more than four patients experienced CR or PR, the study would be considered a success. The trial was conducted from September 1, 2015. If the time of death or progression could not be confirmed or the patient was still alive, we recorded the date of the last follow‐up as censored data for the survival analysis. Clinical variables are described using percentages and median. Time estimation and survival were done by Kaplan‐Meier method. We did all analyses with SPSS software 25.0 (IBM Corporation, Armonk, NY).
Investigator's Analysis	Active and should be pursued further

Drug Information

Bevacizumab
Generic Name	Bevacizumab
Trade Name	Avastin
Company Name	Roche
Drug Type	Antibody
Drug Class	Angiogenesis ‐ VEGF
Dose	7.5 milligrams (mg) per kilogram (kg)
Route	IV
Schedule of Administration	Patients received bevacizumab 7.5 mg/kg on day 1 every 3 weeks
S‐1
Generic Name	S‐1
Trade Name	Tegafur, gimeracil, and oteracil potassium capsules
Company Name	Shandong New Time Pharmaceutical Co., Ltd.
Drug Type	Small molecule
Drug Class	Antimetabolite
Dose	BSA <1.25 m2, 80 mg/day; BSA ≥1.25 m2 but <1.5 m2, 100 mg/day; and BSA ≥1.5 m2, 120 mg/day; milligrams (mg) per flat dose milligrams (mg) per squared meter (m2)
Route	Oral (p.o.)
Schedule of Administration	Patients received S‐1 on days 1–14 every 3 weeks. Dose was based on BSA.
Raltitrexed
Generic Name	Raltitrexed
Trade Name	Raltitrexed
Company Name	Nanjing CHIA TAI Tianqing Pharmaceutical Co. Ltd.
Drug Type	Small molecule
Drug Class	Antimetabolite
Dose	3 milligrams (mg) per squared meter (m2)
Route	IV
Schedule of Administration	Patients received raltitrexed 3 mg/m2 on day 1 every 3 weeks

Patient Characteristics

Number of Patients, Male	26
Number of Patients, Female	18
Stage	Eligibility required pathologic diagnosis confirming metastatic and unresectable colorectal adenocarcinoma
Age	Median (range): 52 (25–76) years
Number of Prior Systemic Therapies	Median (range): 2 (2–3)
Performance Status: ECOG	0 — 15 1 — 24 2 — 5 3 — 0 Unknown — 0
Other	Primary site of disease: right side, 8; left side, 36. Differentiation of tumors: poor, 4; moderate or well, 30; unknown, 10. Prior targeted therapy: cetuximab, 9; bevacizumab, 29. KRAS/NRAS/BRAF status: KRAS mutation, 23; NRAS mutation, 3; BRAF mutation, 2; Unknown, 7.
Cancer Types or Histologic Subtypes	Colorectal adenocarcinoma, 44

Primary Assessment Method

Title	Total Patient Population
Number of Patients Screened	44
Number of Patients Enrolled	44
Number of Patients Evaluable for Toxicity	44
Number of Patients Evaluated for Efficacy	44
Evaluation Method	RECIST 1.1
Response Assessment CR	n = 0 (0%)
Response Assessment PR	n = 7 (15.9%)
Response Assessment SD	n = 17 (38.6%)
Response Assessment PD	n = 20 (45.5%)
Response Assessment OTHER	n = 0 (0%)
(Median) Duration Assessments PFS	110 days, CI: 65.0–155.0
(Median) Duration Assessments OS	367 days, CI: 310.4–423.6
Waterfall plot legend	Waterfall plot of evaluable patients (n = 44) showing the largest decrease in the sum of the target lesions compared with baseline.
Outcome Notes	There were no complete responses; ORR was 15.9%, and the disease control rate was 54.5%.

Adverse Events

All Cycles
Name	NC/NA, %	Grade 1, %	Grade 2, %	Grade 3, %	Grade 4, %	Grade 5, %	All grades, %
Anemia	52	30	14	5	0	0	48
AST, SGOT	59	25	11	5	0	0	41
ALT, SGPT	68	23	9	0	0	0	32
Leukocytes (total WBC)	68	16	11	5	0	0	32
Hypoalbuminemia	75	16	9	0	0	0	25
Neutrophils/granulocytes (ANC/AGC)	80	11	7	2	0	0	20
Platelets	80	14	5	2	0	0	20
Proteinuria	84	11	5	0	0	0	16
Bilirubin (hyperbilirubinemia)	91	5	0	5	0	0	9
Anorexia	64	32	5	0	0	0	36
Diarrhea	73	18	7	2	0	0	27
Fatigue	84	16	0	0	0	0	16
Vomiting	84	14	2	0	0	0	16
Petechiae/purpura (hemorrhage/bleeding into skin or mucosa)	70	30	0	0	0	0	30
Hypertension	82	16	2	0	0	0	18
Hyperpigmentation	59	25	16	0	0	0	41
Pruritus/itching	86	9	2	2	0	0	14

Abbreviations: AGC, absolute granulocyte count; ALT, alanine aminotransferase; ANC, absolute neutrophil count; AST, aspartate transaminase; NC/NA, no change from baseline/no adverse event; SGOT, serum glutamic oxaloacetic transaminase; SGPT, serum glutamic pyruvic transaminase; WBC, white blood cell.

Assessment, Analysis, and Discussion

Completion	Study completed
Investigator's Assessment	Active and should be pursued further

Currently, regorafenib, fruquintinib, and TAS‐102 are standard third‐line treatments for refractory metastatic colorectal cancer (mCRC) [6, 7, 8, 9, 10]. For patients with wild‐type RAS mCRC, cetuximab combined with irinotecan may also be considered. But these regimens show limited efficacy. Our previous study indicated that the S‐1 and raltitrexed combination offered improved objective response rate (ORR) and significant survival benefit in patients with mCRC treated in the third or later line [4]. However, the efficacy and toxicity of bevacizumab combined with S‐1 and raltitrexed remained unknown. In this open‐label, single‐arm, phase II study of patients with mCRC after failure of fluoropyrimidine, irinotecan, and oxaliplatin, bevacizumab combined with S‐1 and raltitrexed demonstrated manageable toxicity and promising antitumor activity. Our study showed a relatively higher ORR of 15.9% (7/44) compared with regorafenib, fruquintinib, and TAS‐102 (all less than 5%), similar median progression‐free survival (mPFS; 3.7 months vs. 1.9–3.7 months), and longer median overall survival (mOS; 12.2 months vs. 5.3–9.3 months) [6, 7, 8, 9, 10]. In our study, about 90% patients only received two prior regimens and enrolled with good performance status. The subsequent use of regorafenib or fruquintinib, thereby increasing the number of lines of therapy, may have contributed to the longer mOS observed in this study. However, this was a single‐center, single‐arm study with small sample size. Larger sample size and randomized controlled studies are needed to fully determine the efficacy of this regimen.

In recent years, sustained anti–vascular endothelial growth factor (VEGF)/VEGF receptor (VEGFR) therapy has become an important management strategy for mCRC after it was shown that bevacizumab treatment needs to be maintained because a discontinuous schedule of bevacizumab in colorectal cancer tends to trigger tumor regrowth and increase tumor resistance [11]. Some researchers concluded that resistance to chemotherapy mainly results from changes of tumor cell biological characteristics and is drug specific [12]. Bevacizumab, on the other hand, is said to be targeted to genetically stable endothelial cells, and the resistance to bevacizumab may be attributed to the development of alternative angiogenesis mechanisms [13]. Therefore, resistance to bevacizumab is unlikely to occur simultaneously with or share the same mechanism as chemotherapy resistance. After chemotherapy resistance, bevacizumab could continue to effectively reduce tumor interstitial pressure, contribute to vascular normalization, and increase the concentration of chemotherapy drug in tumor tissues [13, 14]. Both the ML18147 and BEBYP trials demonstrated that the reintroduction or the continuation of bevacizumab with second‐line chemotherapy beyond first progression was associated with a significantly longer progression‐free survival (PFS) and overall survival (OS) [15, 16]. According to the SPIRITT and PRODIGE18 trials, even in patients with wild‐type RAS mCRC, after bevacizumab combination with chemotherapy in the first line, the mPFS of continuous administration of bevacizumab in the second line was similar to that of chemotherapy in combination with anti–epidermal growth factor receptor monoclonal antibodies, and the OS benefit was more obvious [17, 18]. The above trials all supported continuation of bevacizumab in second‐line therapy after the progression of bevacizumab plus first‐line chemotherapy.

However, little is known about the effects of continuous use of angiogenesis inhibitors combined with chemotherapy in third‐ or later‐line therapy. Last year, a retrospective study firstly reported that bevacizumab in combination with TAS‐102 was significantly associated with improved mPFS and mOS in patients with mCRC refractory to standard therapies [19]. Earlier this year, a randomized, phase II study showed that TAS‐102 plus bevacizumab was associated with an improvement of PFS (4.6 months vs. 2.6 months) and tolerable toxicity compared with TAS‐102 monotherapy in patients with refractory mCRC, which might be a new treatment option for them [20]. It supported the effective outcome of sustained anti‐VEGF/VEGFR therapy in third‐ or later‐line therapy of mCRC. In our current study, bevacizumab combined with S‐1 and raltitrexed showed a higher ORR (15.9% vs. 13.0%), a similar mPFS (3.7 months vs. 3.6 months), and a similar mOS (12.2 months vs. 12.4 months), compared with S‐1 and raltitrexed regimen alone in our previous study, which may provide another new treatment option for refractory mCRC. Whether adding bevacizumab to raltitrexed combined with S‐1 is better than chemotherapy alone needs further randomized controlled studies to confirm.

The mechanism of the synergetic effect between S‐1 and raltitrexed is not well understood. Several basic studies have found upregulation of dihydropyrimidine dehydrogenase (DPD) and thymidylate synthase (TS) in mCRC after exposure to fluorouracil (5‐FU) and/or its analogues [21, 22]. S‐1 contains a DPD inhibitor, and raltitrexed is a TS inhibitor. We speculated that the combination of S‐1 and raltitrexed may play a synergistic inhibitory effect on TS, especially in the condition of DPD and TS upregulation after long‐term exposure to 5‐FU, which may reverse the resistance to 5‐FU [23]. Our preliminary research and this trial showed the efficacy of the S‐1 and raltitrexed combination with or without bevacizumab in refractory mCRC.

As for side effects, bevacizumab combined with S‐1 and raltitrexed was well tolerated, and the observed common toxicity in our trial was similar to that reported in our previous study [4]. The most common adverse events of this regimen in our study were bone marrow depression, hyperpigmentation, dysfunction of digestive system, abnormality of liver function, and bleeding. The most common grade 3/4 toxicities were anemia (4.5%), leukopenia (4.5%), diarrhea (2.3%), and rash (2.3%). No treatment‐related serious adverse events occurred. Combining bevacizumab with S‐1 and raltitrexed did not increase the risk of serious adverse events but only increased the risk of grade 1/2 bleeding, hypertension, proteinuria, and venous thromboembolic events. Compared with TAS‐102 plus bevacizumab, our scheme had less grade 3/4 neutropenia (2.3% vs. 67%) and grade 3/4 diarrhea (2.3% vs. 9%) [20]. Furthermore, the safety profile of our combination is different from antiangiogenic tyrosine kinase inhibitors, such as regorafenib and fruquintinib. The incidences of rash, proteinuria, bilirubin, and hypertension in bevacizumab plus S‐1 and raltitrexed were lower than those in regorafenib and fruquintinib [6, 7, 10].

As far as we know, this trial is the first phase II trial to prospectively evaluate the efficacy and safety of bevacizumab combined with S‐1 and raltitrexed in patients with mCRC refractory to standard therapies. In this study, the combination appeared well tolerated and showed promising antitumor activity in patients with mCRC who experienced disease progression after treatment with fluoropyrimidine, irinotecan, and oxaliplatin. Several limitations were as follows: firstly, this was an investigator‐initiated, single‐arm, prospective phase II trial with a small number of patients, but also, PFS and OS were validated by the investigators and not assessed by an external independent committee. Secondly, at the end of the observation period, 11 patients were censored on the Kaplan‐Meier curve for different reasons, which might bias the PFS and OS results.

In conclusion, bevacizumab combined with S‐1 and raltitrexed shows positive antitumor efficacy and acceptable toxicity and might represent a treatment option for patients with mCRC after failure of fluoropyrimidine, irinotecan, and oxaliplatin. Larger sample size and randomized controlled studies are needed to definitively determine the efficacy of this regimen.

Disclosures

The authors indicated no financial relationships.