Regorafenib in refractory gastric cancer: modest median, meaningful milestone

Gastric cancer/gastroesophageal junction cancer (GC/GEJC) rank fifth in incidence and cancer-related mortality worldwide, and they are among the most common and lethal malignancies (1). Despite considerable progress in first- and second-line settings, the prognosis of patients with advanced GC/GEJC disease remains poor following a failed response to standard therapies (2-6). Over the past decade, several pivotal trials have shaped the treatment landscape for these refractory diseases. The ATTRACTION-2 study established nivolumab as a salvage option for Asian populations. The TAGS trial confirmed the benefit of trifluridine/tipiracil globally, whereas ramucirumab, either alone or in combination with paclitaxel, has become an established second-line treatment (7-9). However, the therapeutic arsenal for patients who do not respond to these treatments is limited. Thus, there is a need for new treatment approaches. Regorafenib is an oral multikinase inhibitor that targets vascular endothelial growth factor receptor (VEGFR), fibroblast growth factor receptor (FGFR), platelet-derived growth factor receptor (PDGFR) and other angiogenic and oncogenic pathways. The phase II INTEGRATE study already suggested an improvement in progression-free survival compared with placebo in previously treated advanced GC/GEJC. These encouraging findings prompted the pivotal phase III INTEGRATE IIa study, which tested regorafenib against placebo in refractory disease (10). With the incorporation of immune checkpoint inhibitors such as nivolumab into first-line chemotherapy regimens, the relevance of later-line nivolumab use—as established in ATTRACTION-2—has become less clear in current practice (3,5,9).

The INTEGRATE IIa was a global, randomized phase III trial comparing regorafenib versus placebo in patients with heavily pretreated advanced gastric or GEJC (10). Among the 251 enrolled patients, 7% had received prior immune checkpoint inhibitors, 13% had human epidermal growth factor receptor 2 (HER2)-positive tumors, and 43% had previous anti-vascular endothelial growth factor (VEGF) treatment such as ramucirumab. The study met its primary endpoint, showing a significant overall survival benefit [median 4.5 vs. 4.0 months; hazard ratio (HR) =0.68; 95 % confidence interval (CI): 0.52–0.90; P=0.006] and longer progression-free survival (HR =0.53; 95% CI: 0.40–0.70). Although the absolute median survival gain was modest, 12-month overall survival improved from 6% to 19%, suggesting durable benefit in a subset of patients. Adverse events were consistent with previous studies and were manageable with dose adjustment and supportive care, supporting feasibility even in heavily pretreated patients (11).

Although cross-trial comparisons should be interpreted with caution, the improvement in 12-month overall survival from 6% to 19% suggests that a subset of patients derive sustained benefit from regorafenib. This long-term survival tail is conceptually similar to patterns observed in salvage immunotherapy trials such as ATTRACTION-2, despite clear differences in mechanism of action and patient populations. These results highlight that median OS alone may not fully capture clinically meaningful benefit in heavily pre-treated disease. Compared with other pivotal trials, regorafenib appears to offer intermediate efficacy. The TAGS trial with trifluridine/tipiracil showed a median OS improvement of 2.1 months (5.7 vs. 3.6 months, HR =0.69), whereas ATTRACTION-2 had a similar HR (0.63), with a higher long-term survival plateau (12-month OS of 26.2%) (Table 1) (7,9). In contrast, the results of the RINDBeRG trial evaluating the addition of ramucirumab to irinotecan were negative, with no overall survival advantage (9.4 vs. 8.5 months, HR =0.92; P=0.49). This highlights the difficulty of improving outcomes with conventional chemotherapy and angiogenesis inhibition (12). Based on these findings, the benefit of regorafenib may appear smaller in median terms; however, it remains meaningful, especially as it was consistently observed across various populations (62% Asian vs. 38% non-Asian). This consistency enhances its generalizability and surpasses the regional limitations of ATTRACTION-2. Moreover, regorafenib is mechanistically distinct from cytotoxic chemotherapy and immunotherapy. As a multikinase inhibitor, it targets VEGFR1–3, FGFR, PDGFR, stem cell growth factor receptor (CD117) (KIT), rearranged during transfection proto-oncogene (RET), and RAF, thereby modulating angiogenesis, stromal signaling, and the tumor microenvironment, and providing a non-cross-resistant option for patients who no longer benefit from standard treatment. This broader kinase inhibition offers a biological rationale for the use of regorafenib after ramucirumab, analogous to other tumor types in which VEGF tyrosine-kinase inhibitors may retain activity following prior therapy with a VEGF monoclonal antibody. This broader inhibition provides a biological rationale for the use of regorafenib after prior treatment with ramucirumab, as sequential VEGF-pathway inhibition through a multikinase inhibitor may retain antitumor activity, similar to other tumor types treated with VEGF monoclonal antibodies followed by tyrosine-kinase inhibitors.

Table 1. Summary of key characteristics and outcomes from pivotal trials for patients with refractory gastric cancer.

Characteristics and outcomes	ATTRACTION-2 (9)	TAGS (7)	RINDBeRG (12)	INTEGRATE IIa (10)
Number	493	507	402	251
Phase	III	III	III	III
Experimental vs. control arm	Nivolumab vs. placebo	FTD/TPI vs. placebo	IRI + Ram vs. IRI	Regorafenib/placebo
Primary endpoint	Overall survival	Overall survival	Overall survival	Overall survival
Treatment line	3rd or later	3rd or later	3rd or later	3rd or later
Asian/non-Asian	100%/0%	15%/85%	100%/0%	62%/38%
Objective response rates	11.2% vs. 0%, Δ+11.2%	4% vs. 2%, Δ+2%	22.1% vs. 15.8%, Δ+6.3%	2.4% vs. 0%, Δ+2.4%
Progression-free survival	1.61 vs. 1.45 months, Δ+0.16 months, HR 0.60, P<0.0001	2.0 vs. 1.8 months, Δ+0.2 months, HR 0.57, P<0.0001	3.8 vs. 2.8 months, Δ+0.4 months, HR 0.73, P=0.002	1.8 vs. 1.6 months, Δ+0.2 months, HR 0.53, P<0.0001
Overall survival	5.26 vs. 4.14 months, Δ+1.12 months, HR 0.63, P<0.001	5.7 vs. 3.6 months, Δ+2.1 months, HR 0.69, P=0.00029	9.4 vs. 8.5 months, Δ+0.9 months, HR 0.92, P=0.49	4.5 vs. 4.0 months, Δ+0.5 months, HR 0.68, P=0.006
12-month overall survival	26.2% vs. 10.9%, Δ+15.2%	21% vs. 13%, Δ+8%	36.5% vs. 32.2%, Δ+4.3%	19% vs. 6%, Δ+13%
Characteristic adverse events	Immune-related adverse events	Myelosuppression, gastrointestinal toxicities	Neutropenia, diarrhea (IRI only)	HFSR, hypertension

FTD/TPI, trifluridine/tipiracil; HFSR, hand-foot skin reaction; HR, hazard ratio; IRI, irinotecan; Ram, ramucirumab.

Despite these strengths, INTEGRATE IIa also has some limitations. The absolute median OS benefit was only half a month, which tempers enthusiasm for regorafenib as a monotherapy. Objective responses were virtually absent, making the drug less attractive for patients with rapidly progressing disease, which require tumor shrinkage. The side effects, while generally manageable, may still be troubling in frail populations; in particular, hand-foot skin reaction, hypertension, fatigue, and liver function abnormalities—observed respectively in 13%, 12%, 8%, and 7% of patients in INTEGRATE IIa—can compromise tolerance and often necessitate dose reductions or interruptions.

Another major limitation of this approach is the absence of predictive biomarkers. Unlike immunotherapy, where programmed death-ligand 1 (PD-L1) expression, microsatellite instability (MSI), or Epstein-Barr virus (EBV) status can guide treatment and stratify patients, regorafenib lacks a validated marker of sensitivity. This affects its integration into precision oncology regimens. Exploratory biomarker experiments in INTEGRATE IIa failed to identify predictors, which underscores the need for additional studies.

Finally, the rapidly evolving treatment landscape raises questions regarding sequencing. With immune checkpoint inhibitors now used in earlier lines, and trifluridine/tipiracil treatment widely adopted, the clinical niche for regorafenib may be narrow. Additional studies are necessary to determine whether regorafenib should precede or follow these agents, or be used to select subgroups. The most promising application for regorafenib may be in combination regimens. The REGONIVO trial, which was a Japanese phase Ib study of regorafenib plus nivolumab, reported robust activity, with response rates exceeding 40% in GC. Although subsequent studies in Western populations were not as impressive, there is still a coherent biological rationale for combining anti-angiogenic therapy with immunotherapy. Anti-angiogenic agents attenuate the tumor microenvironment, enhance T-cell infiltration, and overcome resistance to checkpoint blockade. Based on this rationale, the ongoing INTEGRATE IIb trial is currently evaluating regorafenib in combination with nivolumab in a randomized phase III study. The outcomes of this study may determine whether regorafenib is a modest monotherapy option or the backbone of effective combination regimens. Similarly, studies combining regorafenib with chemotherapy or other targeted agents are under development.

With respect to biomarkers, exploratory analyses using circulating tumor DNA, angiogenic signatures, or immune-related markers may stratify patients for various therapeutic regimens. Identifying subgroups most likely to benefit could amplify regorafenib’s clinical value and justify its use in certain medical situations.

In real-world practice, considerations such as cost, toxicity management, and availability are likely to impact the use of regorafenib. In resource-limited settings, an oral agent with global accessibility has considerable advantages, provided its benefit-to-risk ratio is carefully managed. The INTEGRATE IIa trial demonstrated that regorafenib is active in refractory GC and GEJC. Although the improvement in median OS was modest, the substantial increase in 1-year survival and the preservation of quality of life (QoL) highlight its clinical relevance. The rigor of this trial, including its placebo-controlled design and multinational enrollment, provides sufficient evidence to support regorafenib as part of the therapeutic armamentarium.

The limitations of regorafenib, including its absolute benefit, lack of responses, and absence of biomarkers, indicate that it is not a definitive solution. Rather, it represents a foundation upon which future strategies may be built. The promise of combination therapy and biomarker-driven selection offers hope that regorafenib can emerge as a viable component of chemotherapeutic regimens. Overall, regorafenib provides a meaningful, albeit modest, survival advantage in refractory GC, extending the lives of some patients who otherwise face few options. For clinical researchers, INTEGRATE IIa provides a foundation for new combinations and precision approaches that could transform regorafenib from a modest monotherapy into a key partner in the ongoing battle against GC. Emerging evidence also suggests that tyrosine kinase inhibitors can potentially synergize with immune checkpoint inhibitors. Across other malignancies, several tyrosine kinase inhibitor (TKI)–immune checkpoint inhibitor (ICI) combinations have demonstrated substantial benefit—for example, in hepatocellular and renal cell carcinomas—whereas other studies have failed to improve outcomes, underscoring that this approach is not uniformly successful and that its role in GC remains to be defined. Elucidating optimal combinations and sequencing strategies will be an important future challenge in the era of precision immuno-oncology.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.