Regorafenib in Patients with Antiangiogenic‐Naïve and Chemotherapy‐Refractory Advanced Colorectal Cancer: Results from a Phase IIb Trial

Regorafenib is an oral multikinase inhibitor with antiangiogenic and antitumor effects. This article reports on the efficacy and safety of regorafenib in patients with antiangiogenic therapy‐naïve chemotherapy‐refractory metastatic colorectal cancer.

Abstract

Background.

Regorafenib is a multikinase inhibitor with antiangiogenic effects that improves overall survival (OS) in metastatic colorectal cancer (mCRC) after failure of standard therapies. We investigated the efficacy and safety of regorafenib in antiangiogenic therapy‐naïve chemotherapy‐refractory advanced colorectal cancer.

Patients and Methods.

This single‐center, single‐arm, phase IIb study (NCT02465502) enrolled adults with mCRC whose disease had progressed on, or who were intolerant to, standard therapy, but who were antiangiogenic therapy‐naïve. Patients received regorafenib 160 mg once daily for 3 weeks per 4‐week cycle. The primary endpoint was progression‐free survival (PFS) rate at week 8.

Results.

Of 59 treated patients, almost half had received at least four prior lines of therapy. Patients received a median of 86% of the planned dose. The week 8 PFS rate was 53% (95% confidence interval [CI], 39.1–64.3); median PFS was 3.5 months (95% CI, 1.8–3.6). Median OS was 7.4 months (95% CI, 5.3–8.9). Tumor response (RECIST version 1.1) was 2%, and metabolic response rate (criteria from the European Organisation for Research and Treatment of Cancer) was 41%. The most frequently reported regorafenib‐related grade ≥3 adverse events were hypertension (36%), hand–foot skin reaction (HFSR, 25%), and hypophosphatemia (24%). There were no regorafenib‐related deaths. An exploratory analysis showed that patients with grade ≥2 HFSR had longer OS (10.2 months) with regorafenib treatment versus those with grades 0–1 (5.4 months).

Conclusion.

These findings support the antitumor activity of regorafenib in antiangiogenic‐naïve patients with chemotherapy‐refractory mCRC.

Implications for Practice.

The multikinase inhibitor regorafenib improved overall survival in the phase III CORRECT and CONCUR trials in heavily pretreated patients with treatment‐refractory metastatic colorectal cancer (mCRC). Exploratory subgroup analysis from CONCUR suggested that regorafenib treatment prior to targeted therapy (including bevacizumab) may improve outcomes. In this single‐center, single‐arm phase IIb study, regorafenib demonstrated antitumor activity in 59 antiangiogenic‐naïve patients with chemotherapy‐refractory mCRC. Further studies should assess the efficacy of regorafenib in this patient population, as well as explore the reasons behind improved outcomes among patients who had a metabolic response and those who developed hand–foot skin reaction.

Introduction

For patients with metastatic colorectal cancer (mCRC) eligible for intensive therapy, chemotherapy based on 5‐fluorouracil and leucovorin with oxaliplatin (FOLFOX) or irinotecan (FOLFIRI), and vascular endothelial growth factor (VEGF) or endothelial growth factor receptor (EGFR) inhibitors, are recommended first‐line systemic therapies [1], [2], [3].

Regorafenib is an oral multikinase inhibitor with antiangiogenic and antitumor effects [4], [5] that is approved to treat mCRC, advanced gastrointestinal stromal tumors, and hepatocellular carcinoma (HCC) [6], [7], [8], [9], [10]. In the phase III randomized CORRECT trial, regorafenib significantly, albeit modestly, improved overall survival (OS) versus placebo in patients with treatment‐refractory mCRC (median OS, 6.4 vs. 5.0 months, respectively; one‐sided p = .0052) [8]. Approximately half of all patients in CORRECT had received at least four prior lines of therapy for mCRC, and all had received prior antiangiogenic therapy with the VEGF inhibitor bevacizumab. The survival benefit with regorafenib versus placebo was confirmed in Asian patients with treatment‐refractory mCRC in the randomized phase III CONCUR trial [11] (median OS, 8.8 vs. 6.3 months, respectively; one‐sided p = .00016). In the CONCUR regorafenib arm, 38% of patients had received at least four lines of therapy for mCRC and, because prior bevacizumab treatment was not mandatory, 41% had not received prior therapy with any targeted agent. An exploratory subgroup analysis suggested that patients who had not received prior targeted therapy (including bevacizumab) had better outcomes with regorafenib than those who had received at least one prior targeted agent [11]. These results, together with more recent evidence [12], suggest that regorafenib treatment prior to targeted therapy may further improve outcomes in patients with mCRC. The current study was designed to investigate the efficacy and safety of regorafenib in patients with antiangiogenic therapy‐naïve chemotherapy‐refractory mCRC.

Materials and Methods

Study Design

In this single‐center, single‐arm, open‐label, phase IIb study (NCT02465502), patients were recruited from a public cancer center in São Paulo, Brazil (Instituto do Câncer do Estado de São Paulo, Universidade de São Paulo). Patients aged at least 18 years with mCRC whose disease had progressed (<4 months) on, or who were intolerant to, standard chemotherapy (based on fluoropyrimidine, oxaliplatin, irinotecan, and anti‐EGFR therapy if RAS wild‐type) were enrolled. Progression was radiologically measured as per the investigator's decision. Patients had to have measurable disease according to Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST v1.1) [13] and an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. Patients who had received prior bevacizumab or another antiangiogenic therapy, or had received any anticancer therapy fewer than 4 weeks from the start of study treatment, were excluded. No antiangiogenic agents are reimbursed by the Brazilian health public system, and therefore bevacizumab was not available to patients with mCRC.

Patients received oral regorafenib 160 mg once daily for 3 weeks on and 1 week off in each 4‐week cycle and continued treatment until disease progression, death, unacceptable toxicity, withdrawal of consent, or investigator decision to withdraw the patient. Regorafenib could be delayed or discontinued at the investigator's discretion. Dose modifications and interruptions because of toxicity were made according to protocol‐specified recommendations.

The study was conducted in accordance with the Declaration of Helsinki, Good Clinical Practice guidelines, and applicable local laws. Approval was obtained from the appropriate ethics committee or institutional review board. All participants provided written informed consent.

Assessments

The primary endpoint was progression‐free survival (PFS) rate at week 8 (percentage of evaluable patients without progression or death by any cause at week 8 of treatment). Secondary endpoints included PFS, OS, overall response rate (ORR), disease control rate (DCR), tumor metabolic response, and safety. Exploratory analyses included evaluating the association between tumor response by RECIST v1.1 [13] and tumor metabolic response by the European Organisation for Research and Treatment of Cancer (EORTC) criteria at week 8 [14], and the association between OS and grade 0–1 or ≥2 adverse events (AEs), specifically hand–foot skin reaction (HFSR), hypophosphatemia, and hypertension.

Tumor response (PFS, ORR, and DCR), using computed tomography (CT) or magnetic resonance imaging (MRI), was assessed by RECIST v1.1 every 8 weeks [13]. Response rates according to RECIST v1.1 were analyzed from conventional contrast‐enhanced CT or MRI scans for all patients. Tumor metabolic response (complete metabolic response [CMR], partial metabolic response [PMR], stable metabolic disease, or progressive metabolic disease), as assessed by [¹⁸F]‐fluorodeoxyglucose (FDG) positron emission tomography/CT (PET/CT), was determined using the EORTC criteria, using percentage change from baseline at week 8 in maximum standardized uptake value (SUV_max) [14]. The FDG‐PET/CT scans were of diagnostic quality with a standard radiation dose but without intravenous contrast and were performed according to the local Nuclear Medicine Service standard procedure. Up to ten lesions with the highest FDG uptake were defined as target lesions at baseline, up to three lesions per organ were included as target lesions, and bone lesions could be included as target lesions. The minimum dimension of target lesions was 10 mm. Positive lesions were required to have a SUV_max >2.5. The same target lesions were measured by PET/CT scans at baseline (within 4 weeks of regorafenib treatment) and at 8 weeks (±5 days). All PET/CT scans were performed at the same institution, using the same protocol, and were read by the same radiologist. The association between tumor response by RECIST v1.1 and tumor metabolic response using the EORTC criteria was analyzed descriptively using the sum of the longest diameter (SLD) as per RECIST v1.1 and SUV_max by FDG‐PET/CT.

Safety assessments were performed at baseline, during the treatment period, end of treatment, and 30–35 days after the last dose, summarized using the Medical Dictionary for Regulatory Activities version 16.0 terms, and graded using the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03.

Statistical Analysis

The efficacy analysis included all patients assigned to study treatment, and the safety analysis included all patients who received at least one dose of study drug.

The primary endpoint, PFS rate at week 8, was tested against the null hypothesis that the true PFS rate at 8 weeks was ≤30%, based on the fact that the study population had potentially more advanced disease than in the CORRECT trial. Based on a one‐sided α error of 5% and a ß error of 10%, 53 evaluable patients were needed, and assuming a screening failure rate of 20% and a dropout rate of 10%, approximately 74 patients would need to be screened and 59 patients enrolled.

OS and PFS were estimated by the Kaplan‐Meier method. Estimates of OS in patients with grade 0–1 or ≥2 AEs (HFSR, hypophosphatemia, and hypertension) were demonstrated using the Kaplan‐Meier method.

Results

Of 74 patients assessed for eligibility between July 21, 2015, and December 11, 2017, 59 were eligible and received at least one dose of regorafenib. The remaining 15 patients were considered ineligible because they failed one or more of the exclusion criteria, primarily because of laboratory abnormalities such as renal or hepatic dysfunction or ECOG status. Reasons for discontinuation were progressive disease (radiologic: n = 52, 88%; clinical: n = 5, 9%), AEs not associated with clinical disease progression (n = 1, 2%), and death from sepsis not considered related to study drug (n = 1, 2%).

The median age was 58 years (range, 30–74), and almost half of patients had received at least four prior lines of therapy on or after diagnosis of metastatic disease; the median time from diagnosis of metastatic disease to regorafenib treatment was 23 months (range 5–75; supplemental online Table 1).

The median regorafenib dose administered was 147.7 mg per day, and patients received a median of 86% of the planned dose. Overall, 92% of patients received a daily regorafenib dose of 120–160 mg. A median of four cycles (range, 1–10) of treatment was administered, and the median duration of regorafenib treatment (including interruptions or delays and drug holidays) was 3.5 months (range, 0.7–9.7). Dose modifications, mostly due to AEs, were required by 46 of 59 patients (78%); 35 patients (59%) required dose reduction and 42 patients (71%) required interruption or delay. After completion of regorafenib treatment, 16 patients (27%) received at least one further systemic treatment, including 5‐fluorouracil (n = 9), folinic acid (n = 9), oxaliplatin (n = 8), capecitabine (n = 2), irinotecan (n = 2), and other (n = 4).

Efficacy

The week 8 PFS rate was 53% (95% confidence interval [CI], 39.1–64.3) and median PFS was 3.5 months (95% CI, 1.8–3.6; Fig. 1A). The week 8 OS rate was 98% (95% CI, 88.6–99.8), and the median OS was 7.4 months (95% CI, 5.3–8.9; Fig. 1B). One patient had a partial response by RECIST v1.1, and 29 had stable disease, achieving an ORR and DCR of 2% and 51%, respectively. Metabolic response rate by the EORTC criteria was 41% (0% CMR, 41% PMR; Table 1). Using percentage change from baseline at week 8, reductions in metabolic activity (using SUV_max) were larger than reductions in tumor volume (using SLD; supplemental online Fig. S1). Furthermore, metabolic responders at week 8 had better PFS and OS than metabolic nonresponders (Fig. 2). In supplemental online Figure 2, baseline and week 8 FDG‐PET/CT scans show a reduction in FDG uptake in the central area of the liver lesion (PMR), compared with contrast‐enhanced CT, which shows stable disease.

Figure 1.

Kaplan‐Meier estimates of survival after regorafenib treatment. (A): Progression‐free survival. (B): Overall survival.

Abbreviations: CI, confidence interval; OS, overall survival; PFS, progression‐free survival.

Table 1. Correlation between tumor response (RECIST version 1.1) and metabolic response (EORTC criteria) categories at week 8 of regorafenib treatment (N = 59).

Figure 2.

Kaplan‐Meier estimates of survival for metabolic responders (CMR+PMR) versus nonresponders (PMD+SMD) at week 8 of regorafenib treatment using EORTC criteria. (A): Progression‐free survival. (B): Overall survival.

^aSix patients were not evaluable (five patients had missing scans and one patient died).

Abbreviations: CI, confidence interval; CMR, complete metabolic response; EORTC, European Organisation for Research and Treatment of Cancer; NA, not available; OS, overall survival; PFS, progression‐free survival; PMD, progressive metabolic disease; PMR, partial metabolic response; SMD, stable metabolic disease.

Safety

All patients experienced at least one treatment‐emergent adverse event (TEAE) and at least one regorafenib‐related TEAE. The most frequently reported TEAEs (any cause and any grade) were HFSR (68%), decreased appetite (63%), and fatigue (66%). The most frequently reported regorafenib‐related TEAEs of any grade were HFSR (68%), fatigue (58%), and decreased appetite (51%; Table 2). Overall, 76% of patients experienced a grade ≥3 regorafenib‐related TEAE, the most common of which were hypertension (36%), HFSR (25%), and hypophosphatemia (24%; Table 2). There were no regorafenib‐related deaths.

Table 2. Summary of treatment‐emergent adverse events.

Medical Dictionary for Regulatory Activities version 16.0 terms with Common Terminology Criteria for Adverse Events version 4.03 grading.

^aAn adverse event (AE) is classified as serious if it results in death, is life‐threatening, requires inpatient hospitalization or prolongation of existing hospitalization, results in persistent or significant disability or incapacity, is a congenital anomaly or birth defect, or is medically important as judged by the investigator. The following hospitalizations are not serious AEs: the admission results in a hospital stay of <12 hours, or the admission is preplanned, or the admission is not associated with an AE. The five serious AEs considered to be related to study drug were myocardial infarction, fatigue, bronchial infection, enterocolitis infectious, and sepsis.

^bDose modifications include delays, interruptions, and reductions.

^cTEAEs reported in ≥25% of patients, regardless of causality and grade, for each preferred term.

Abbreviation: TEAE, treatment‐emergent adverse event.

Twenty‐six patients (44%) had dose reductions as a result of an AE, most frequently due to HFSR (n = 14, 24%) and hypertension (n = 4, 7%). Thirty‐nine patients (66%) had dose interruptions because of an AE; the most common reasons were HFSR (n = 16, 27%) and hypertension (n = 14, 24%). One patient (2%) discontinued treatment because of grade 3 regorafenib‐related HFSR after 25 weeks of treatment (7 cycles).

Analysis of Adverse Events and Survival

An exploratory analysis of specific AEs and OS showed that patients receiving regorafenib who developed grade ≥2 HFSR had longer OS than patients who developed grade 0–1 HFSR (10.2 vs. 5.4 months, respectively; supplemental online Fig. 3). No difference was observed for hypophosphatemia and hypertension (supplemental online Figs. 4 and 5).

Discussion

In this single‐arm, phase IIb study, the week 8 PFS rate of 53%, median PFS of 3.5 months, and median OS of 7.4 months support the antitumor activity of regorafenib in antiangiogenic therapy‐naïve patients with chemotherapy‐refractory advanced colorectal cancer. This is important as this was a heavily pretreated population of patients (48% had received at least four prior lines of therapy in the metastatic setting) with an aggressive tumor profile (disease progression <4 months of study entry, similar to the 3 months in CORRECT and CONCUR). In the current study, 27% of patients were able to receive further systemic therapy after regorafenib (including 5‐fluorouracil, folinic acid, and oxaliplatin), similar to CORRECT (26%) and CONCUR (31%) [8], [11]. Interestingly, patients who experienced grade ≥2 HFSR and/or metabolic response by PET/CT had longer median OS (10.2 and 8.5 months, respectively) than the overall cohort.

The impact of prior targeted therapy, including antiangiogenic agents, on the efficacy of regorafenib has been explored in the phase III CONCUR trial, in which 60% of patients had received prior targeted therapy and only 41% had received prior bevacizumab [11]. An exploratory analysis of CONCUR suggested that patients who had not received prior targeted therapy (anti‐VEGF or anti‐EGFR) had a greater benefit compared with patients who had received at least one prior targeted therapy [11]. The authors speculated that the efficacy of regorafenib may be affected by prior targeted therapy, which may explain the apparently larger OS benefit observed in CONCUR versus CORRECT (8.8 vs. 6.4 months, respectively), a phase III trial in which all patients had received prior bevacizumab [8], [11]. More recently, a randomized, phase II study (REVERCE) reported a greater OS benefit for patients with EGFR inhibitor‐naïve mCRC who received regorafenib followed by cetuximab than vice versa, although it should be noted that most of these patients had received prior bevacizumab [12]. In addition, in a regorafenib expanded access program conducted in our institution 1 year earlier, OS was 6.2 months among the 24 patients with chemotherapy‐refractory mCRC previously treated with bevacizumab or aflibercept, according to an independent analysis at our institution [15]. However, outcomes in the phase III VELOUR trial, which evaluated the VEGF inhibitor aflibercept in combination with chemotherapy in second‐line mCRC, were not affected by prior bevacizumab [16], which may be partly due to differences in the target profile of aflibercept (inhibitor of VEGFA, VEGFB, and placental growth factor) and regorafenib (a multikinase inhibitor).

Differences between tumor response assessments using RECIST v1.1 and EORTC criteria, with a greater proportion of patients having a response using the EORTC criteria, suggest that metabolic response assessment may be a better predictor of antitumor activity and OS in patients receiving antiangiogenic agents, and we observed that metabolic responders lived longer than metabolic nonresponders. The reason for this difference is that RECIST measures changes in tumor size and does not detect metabolic or vasculature changes, which often occur as a result of treatment with tyrosine kinase inhibitors (TKIs), especially antiangiogenic agents [17], [18]. FDG‐PET/CT imaging provides more details of functional and metabolic changes in lesions [19], [20], with several metabolic response criteria now proposed in place of RECIST [17], [21], [22].

The safety profile of regorafenib was generally consistent with that previously reported in CORRECT and CONCUR, although there were some notable differences [8], [11]. There was a higher incidence of hypophosphatemia in the current study, as well as hypertension, perhaps because these patients were receiving a TKI for the first time. Hypophosphatemia was mostly asymptomatic (and phosphate replacement was not performed), although this may have contributed to the high incidence of fatigue reported. The incidence of regorafenib‐related HFSR (all grades and grade ≥3) was also slightly higher than in CORRECT, which may relate to differences in footwear quality or foot care practices between populations. Regorafenib‐related diarrhea was reported less frequently in this study than in CORRECT, perhaps related to dietary differences between populations. The relationship between HFSR and OS with regorafenib in this study confirms the findings of retrospective analyses from CORRECT in mCRC and RESORCE in HCC, in which patients who had HFSR had a greater treatment benefit with regorafenib than those without HFSR [23], [24]. Although the development of HFSR has been associated with better outcomes with regorafenib and other multikinase inhibitors, oncologists should not pursue the occurrence of HFSR by increasing regorafenib dose intensity, but instead should follow the recommended dose modifications for grade 2 HFSR.

Only one patient in the current study discontinued treatment because of a TEAE (HFSR), in part because of intensive patient monitoring (weekly visits and patients could call at any time) and the prompt management of TEAEs. More than 90% of patients remained on a regorafenib dose between 120 and 160 mg per day, reinforcing that this daily dose range is tolerable and effective. However, data from a recent study demonstrated that patients can also benefit from weekly dose escalation of regorafenib from 80 mg per day to 160 mg per day, in terms of remaining on treatment [25].

Limitations of this study include that this was a single‐center, single‐arm study with a small sample size, and that no correlative studies such as pharmacokinetics of soluble angiogenic factors were performed. However, it was conducted using the same eligibility criteria and ethical and quality standards as in the registrational controlled trials, and the outcomes were consistent with those from previous randomized, double‐blind, phase III trials of regorafenib.

Conclusion

Overall, findings from this study and others suggest that regorafenib has antitumor activity in antiangiogenic‐naïve patients with chemotherapy‐refractory advanced CRC. Randomized trials are required to assess the efficacy of regorafenib in this patient population, as well as explore the reasons behind improved outcomes among patients who had a metabolic response and those who developed HFSR.

Data Sharing Statement

Availability of the data underlying this publication will be determined according to Bayer's commitment to the EFPIA/PhRMA “Principles for responsible clinical trial data sharing.” This pertains to scope, time point, and process of data access. As such, Bayer commits to sharing upon request from qualified scientific and medical researchers patient‐level clinical trial data, study‐level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the U.S. and European Union (EU) as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and U.S. regulatory agencies on or after January 01, 2014. Interested researchers can use www.clinicalstudydatarequest.com to request access to anonymized patient‐level data and supporting documents from clinical studies to conduct further research that can help advance medical science or improve patient care. Information on the Bayer criteria for listing studies and other relevant information is provided in the “Study sponsors section” of the portal. Data access will be granted to anonymized patient‐level data, protocols, and clinical study reports after approval by an independent scientific review panel. Bayer is not involved in the decisions made by the independent review panel. Bayer will take all necessary measures to ensure that patient privacy is safeguarded.

See http://www.TheOncologist.com for supplemental material available online.

Author Contributions

Conception/design: Rachel P. Riechelmann, Paulo M. Hoff

Provision of study material or patients: Rachel P. Riechelmann, Luiz S. Leite, Giovanni M. Bariani, Joao Glasberg, Thomas G. Rivelli, Leonardo Gomes da Fonseca, Daniela R. Nebuloni, Maria I. Braghiroli, Marcelo A. Queiroz

Collection and/or assembly of data: Rachel P. Riechelmann, Luiz S. Leite, Giovanni M. Bariani, Joao Glasberg, Thomas G. Rivelli, Leonardo Gomes da Fonseca, Daniela R. Nebuloni, Maria I. Braghiroli, Marcelo A. Queiroz

Data analysis and interpretation: Rachel P. Riechelmann, Alice M. Isejima, Christian Kappeler, Luciana Kikuchi

Manuscript writing: Rachel P. Riechelmann, Luiz S. Leite, Giovanni M. Bariani, Joao Glasberg, Thomas G. Rivelli, Leonardo Gomes da Fonseca, Daniela R. Nebuloni, Maria I. Braghiroli, Marcelo A. Queiroz, Alice M. Isejima, Christian Kappeler, Luciana Kikuchi, Paulo M. Hoff

Final approval of manuscript: Rachel P. Riechelmann, Luiz S. Leite, Giovanni M. Bariani, Joao Glasberg, Thomas G. Rivelli, Leonardo Gomes da Fonseca, Daniela R. Nebuloni, Maria I. Braghiroli, Marcelo A. Queiroz, Alice M. Isejima, Christian Kappeler, Luciana Kikuchi, Paulo M. Hoff

Disclosures

Rachel P. Riechelmann: Bayer, Roche, Amgen (RF), Bayer (H), Bayer, Servier (SAB); Leonardo Gomes da Fonseca: Bayer, Ipsen (Other: travel grants); Daniela R. Nebuloni: Bayer (Other: travel grants); Maria I. Braghiroli: Roche, Merck Sharp & Dohme (H), Merck (SAB); Alice M. Isejima: Bayer (E); Christian Kappeler: Bayer Bayer (E); Luciana Kikuchi: Bayer (E); Alice M. Isejima: Bayer (E). The other authors indicated no financial relationships.

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board