Rethinking the timing of adjuvant chemotherapy in advanced gastric cancer: From feasibility to precision sequencing

Abstract

The optimal timing for initiating adjuvant chemotherapy (AC) in stage II/III gastric cancer (GC) remains a critical and unresolved clinical issue. Current guidelines lack definitive recommendations, often relying on traditional paradigms that prioritize postoperative recovery over tumor biology. A recent retrospective study challenged this status quo by evaluating ultra-early AC (initiated 10-13 days postoperatively) within Enhanced Recovery After Surgery (ERAS) protocols, confirming its feasibility and safety with a hypothesis-generating signal of reduced peritoneal recurrence, though no survival benefit was observed. This review critically analyzes the study’s findings and limitations, which include the absence of molecular subtyping, inadequate dose optimization, and a small sample size. We propose a paradigm shift towards “precision timing”, an approach that integrates three core pillars: Tumor biology (including molecular subtypes and circulating tumor DNA dynamics), patient recovery status (leveraging ERAS metrics), and dynamic treatment monitoring. By synthesizing evidence from postoperative immunosuppression biology, the unique pathophysiology of peritoneal recurrence in GC, and emerging biomarkers, we delineate a roadmap for future research. This includes the need for prospective trials with enriched high-risk subgroups, biomarker-integrated adaptive designs, and standardized protocols for dose optimization and endpoint adjudication. Moving from a “one-size-fits-all” timing strategy to a biologically informed, individualized approach is essential to improve outcomes for patients with resected GC.

Key Words: Adjuvant chemotherapy; Gastric cancer; Timing of chemotherapy; Precision oncology; Peritoneal recurrence; Circulating tumor DNA; Enhanced Recovery After Surgery

Core Tip: The optimal timing for adjuvant chemotherapy (AC) in gastric cancer (GC) is undefined. A previous study demonstrated that ultra-early AC (10-13 days post-surgery) is feasible and safe under Enhanced Recovery After Surgery protocols, with a signal for reduced peritoneal recurrence but no survival advantage. The study’s limitations, including a lack of molecular profiling and small sample size, underscore the need for a more sophisticated approach. We propose a “precision timing” framework that integrates tumor biology (e.g., molecular subtypes and circulating tumor DNA), patient recovery, and dynamic monitoring to guide AC initiation. This shift from a traditional feasibility paradigm to biomarker-driven, individualized sequencing holds the key to improving outcomes in high-risk GC.

INTRODUCTION

Gastric cancer (GC) remains a leading cause of cancer-related mortality worldwide, with stage II/III disease exhibiting a substantial risk of recurrence despite curative-intent gastrectomy and adjuvant chemotherapy (AC)[1-3]. While the survival benefit of AC over surgery alone is well-established[4-7], a critical and unresolved question is the optimal timing for its initiation. Current clinical guidelines from major bodies such as the National Comprehensive Cancer Network and the Chinese Society of Clinical Oncology provide no definitive recommendation, typically advising initiation within 6-8 weeks postoperatively, a paradigm rooted in prioritizing surgical recovery and patient safety over the potential biological imperatives of early micrometastatic disease control[8-10].

This traditional approach is increasingly being challenged by a growing body of evidence from other gastrointestinal malignancies, such as colorectal and pancreatic cancers, suggesting that early AC initiation may be associated with improved survival[11-14]. The rationale for such a shift in GC is particularly compelling. The postoperative period is characterized by a transient but profound state of immunosuppression, often termed the “immunosuppressive window” (1-7 days postoperatively), which may facilitate the survival and proliferation of residual micrometastases and circulating tumor cells[15-18]. Furthermore, GC exhibits a uniquely high propensity for peritoneal recurrence, which often originates from intraoperative tumor cell shedding, a process that could theoretically be targeted by early systemic therapy[19-22].

In this context, the recent retrospective study by Lin et al[23] represents a pivotal contribution. As the first study to systematically evaluate ultra-early AC (initiated 10-13 days postoperatively) in GC under Enhanced Recovery After Surgery (ERAS) protocols, it provides crucial feasibility and safety data, while also generating a hypothesis regarding its potential to reduce peritoneal recurrence[24]. This opinion review aims to critically analyze the study by Lin et al[23], placing its findings within the broader context of the ongoing debate on AC timing. We will dissect its methodological strengths and limitations, explore the underlying biological rationale for early AC, and ultimately propose a new conceptual framework—”precision timing”—that moves beyond a binary “early vs late” debate towards a personalized, biomarker-guided strategy for treatment sequencing in resected GC.

MAIN FINDINGS AND CRITICAL APPRAISAL OF THE STUDY

Core findings

Lin et al[23] conducted a retrospective cohort study including 68 GC patients who underwent curative gastrectomy. The patients were stratified into an ultra-early AC group (n = 21), receiving chemotherapy 10-13 days postoperatively, and a conventional timing group (n = 47), receiving AC 4-6 weeks postoperatively. The study yielded three key findings: (1) Feasibility and safety: Ultra-early AC was feasible and safe under ERAS protocols. The rates of grade 3/4 toxicity were comparable between the ultra-early and conventional groups (42.9% vs 57.1%, P = 0.285), although the ultra-early group had a significantly higher rate of dose reductions (57.1% vs 26.2%, P = 0.026); (2) Signal of reduced peritoneal recurrence: A hypothesis-generating signal of reduced peritoneal recurrence was observed in the ultra-early group (4.8% vs 26.2%, P = 0.048 by Fisher’s exact test). However, this trend was not statistically significant in the multivariate Cox regression analysis (hazard ratio = 0.418, P = 0.257); and (3) No survival benefit: The 3-year recurrence-free survival (53.7% vs 61.6%) and overall survival (69.1% vs 66.3%) were similar between the two groups. The authors suggested that the lack of survival benefit might be potentially offset by a lower relative dose intensity (RDI) in the ultra-early group (67.5% vs 73.4%).

Methodological limitations and interpretive challenges

While the study by Lin et al[23] is commendable for its pioneering focus, its conclusions must be interpreted within the context of several significant limitations, which also inform future research directions: (1) Study design and sample size: The retrospective, non-randomized design is inherently susceptible to selection bias. More critically, the sample size is extremely small (n = 21 in the ultra-early group), which severely limits statistical power. A post hoc power calculation for overall survival would likely be well below 80%, making the study underpowered to detect clinically meaningful survival differences; (2) Lack of molecular profiling: The study treated GC as a homogeneous disease, ignoring its well-documented molecular heterogeneity as defined by The Cancer Genome Atlas and the Asian Cancer Research Group[25-28]. Subtypes such as microsatellite instability-high (MSI-H), Epstein-Barr virus-positive (EBV⁺), and chromosomal instability (CIN) tumors have distinct biology and may respond differently to chemotherapy and immune modulation[29-32]. The potential differential benefit of ultra-early AC across these subtypes remains unexplored; (3) Dosing and protocol heterogeneity: The study did not provide a clear definition of dose adjustment criteria, tolerance assessment protocols, or dose re-escalation rules, making it difficult to replicate the dosing strategy. The lower RDI observed in the ultra-early group raises a critical question: Is the potential benefit of earlier timing offset by reduced treatment intensity? Future studies must prioritize the optimization of dosing schedules to maintain therapeutic intensity[33]; (4) Outcome assessment and confounding: The diagnosis of peritoneal recurrence was not standardized with blinded adjudication, introducing potential subjective bias. Important confounders, including tumor grade, lymphovascular invasion, Lauren classification, nutritional status, and specific ERAS protocol components (e.g., nutritional support and pain management), were not accounted for in the analysis; and (5) Biological justification: The selection of the 10- to 13-day window as “ultra-early” lacks a clear biological rationale. While it avoids the immediate postoperative immunosuppressive nadir, its connection to the dynamics of micrometastatic disease and the peritoneal tumor microenvironment is not established.

TOWARD PRECISION TIMING: A NEW PARADIGM FOR AC SEQUENCING

The limitations of the study by Lin et al[23] highlight a crucial point: Framing the AC timing question as a simple “early vs late” binary is an oversimplification. We propose a paradigm shift towards precision timing, a strategy that leverages biological insights and dynamic monitoring to individualize the initiation of AC (Figure 1). This approach is built on three core pillars: Tumor biology, patient recovery, and dynamic monitoring.

Figure 1 The precision timing framework for adjuvant chemotherapy in gastric cancer. This conceptual framework integrates three core pillars to guide individualized timing of adjuvant chemotherapy: (1) Tumor biology: Molecular subtyping (e.g., microsatellite instability, Epstein-Barr virus, and chromosomal instability) and detection of postoperative circulating tumor DNA (ctDNA) define the risk of recurrence and potential therapeutic sensitivity; (2) Patient recovery: Objective metrics from Enhanced Recovery After Surgery protocols, including functional recovery (QoR-15) and nutritional status, determine patient readiness for early chemotherapy; and (3) Dynamic monitoring: Real-time assessment of ctDNA clearance and immune status allows for adaptive treatment decisions, including dose adjustment or treatment de-escalation. The intersection of these pillars enables a shift from a fixed-timing paradigm to a biologically informed, personalized treatment sequence. MSI: Microsatellite instability; EBV: Epstein-Barr virus; CIN: Chromosomal instability; ctDNA: Circulating tumor DNA; ERAS: Enhanced Recovery After Surgery; AC: Adjuvant chemotherapy.

Biological basis for timing optimization

The postoperative period is not a static state of recovery but a dynamic phase of altered physiology that can influence tumor behavior[15,17]. The “immunosuppressive window” (1-7 days post-surgery) is characterized by a surge in circulating myeloid-derived suppressor cells and regulatory T cells (Tregs), alongside a decrease in natural killer cell activity and T-cell function[34-37]. This environment paradoxically promotes the survival and engraftment of residual tumor cells and circulating tumor cells, which are often shed during surgery[38-40]. This biological rationale suggests that the “window of opportunity” to target micrometastases might be during or immediately after this immunosuppressive nadir.

GC’s unique pattern of recurrence—predominantly peritoneal—further strengthens this rationale. Peritoneal dissemination often arises from tumor cells exfoliated during surgery, which then implant and proliferate on the peritoneal surface[19,41,42]. Theoretically, ultra-early systemic therapy could target these cells before they establish a stable niche in the peritoneum, a concept supported by some preclinical models[43]. The implementation of ERAS protocols, which reduce surgical stress and accelerate physiological recovery, is a crucial enabler for precision timing. By minimizing postoperative ileus, maintaining nutritional status, and reducing inflammatory stress, ERAS creates a more favorable environment for early AC initiation[44-46].

Molecular heterogeneity and subgroup targeting

The future of AC timing lies in identifying which patients are most likely to benefit from an ultra-early strategy. This requires moving beyond clinicopathological staging to incorporate molecular biomarkers.

Molecular subtypes: MSI-H tumors, which are known to be less chemosensitive, might derive limited benefit from ultra-early cytotoxic AC and may instead be better suited for perioperative immunotherapy[29,47,48]. Conversely, CIN or EBV⁺ tumors, which are often more proliferative and immunogenic, might be more vulnerable to early chemotherapy and immune activation[25,31,49].

Circulating tumor DNA: Postoperative circulating tumor DNA (ctDNA) detection is a powerful prognostic marker that identifies patients with minimal residual disease (MRD) who are at the highest risk of recurrence[50-53]. The presence of ctDNA after surgery provides a compelling rationale for immediate, ultra-early AC to eradicate MRD. Conversely, ctDNA-negative patients may be spared the toxicity of early AC, allowing for a longer recovery period and potentially de-escalated treatment. This “ctDNA-guided” approach is a cornerstone of precision timing[54-56].

Peritoneal recurrence risk markers: Biomarkers associated with peritoneal recurrence, such as elevated levels of cytokines (e.g., interleukin-6 and tumor necrosis factor-alpha) or specific microRNAs (e.g., miR-21) in peritoneal fluid, could help identify patients who would most benefit from early intraperitoneal or systemic therapy[57-59].

The role of dynamic monitoring and patient status

Precision timing requires a dynamic, not static, approach. The decision of when to start AC should be continuously informed by a patient’s real-time status.

Patient recovery: Objective measures of recovery, such as the Quality of Recovery (QoR-15) score, functional capacity, and nutritional markers (e.g., prealbumin), are critical[60]. A patient with rapid recovery under ERAS may be an ideal candidate for ultra-early AC, while one with significant complications may require a delayed start.

Adaptive dosing: The challenge of lower RDI in ultra-early arms can be addressed by personalized dosing strategies. This includes weight-based dosing, tolerance-guided dose adjustments, and the use of growth factor support to maintain dose intensity[61]. Future studies must define clear criteria for dose modifications and re-escalation to optimize the therapeutic window.

FUTURE RESEARCH PRIORITIES AND CLINICAL IMPLICATIONS

The transition from feasibility to precision timing demands a new generation of clinical trials and translational research (Figure 2). The key priorities are listed below.

Figure 2 Proposed roadmap for future research on precision timing in gastric cancer. This schematic outlines the key steps and priorities required to transition from feasibility studies to clinical implementation. The roadmap is divided into three phases: (1) Discovery & validation: Focuses on mechanistic studies to define the optimal biological window, identification and validation of robust biomarkers [e.g., circulating tumor DNA (ctDNA) and immune profiles], and the establishment of standardized Enhanced Recovery After Surgery protocols; (2) Clinical development: Encompasses prospective, biomarker-enriched randomized controlled trials with adaptive designs to evaluate the efficacy of ctDNA-guided timing, alongside dose optimization studies to maintain relative dose intensity; and (3) Implementation & standardization: Involves the development of clinical guidelines, incorporation of patient-reported outcomes and cost-effectiveness analyses, and the establishment of standardized, blinded endpoints for trials. ERAS: Enhanced Recovery After Surgery; ctDNA: Circulating tumor DNA; RCT: Randomized controlled trial; RDI: Relative dose intensity.

The community must move beyond retrospective studies and prioritize large, prospective randomized controlled trials designed with pre-specified power calculations and effect-size targets. Most importantly, these trials must incorporate molecular stratification (e.g., by microsatellite instability and ctDNA status) to identify high-risk subgroups most likely to benefit. Adaptive trial designs that allow for dynamic treatment allocation based on ctDNA clearance are particularly attractive.

Dose optimization and regimen standardization

Future trials must mandate clear protocols for dose management, including the reporting of RDI, criteria for dose reduction, and algorithms for dose re-escalation. The goal is to ensure that any benefit from earlier timing is not negated by a compromise in therapeutic intensity.

Integration of ERAS and nutrition

ERAS protocols and perioperative nutritional support should be standardized and described in detail to ensure reproducibility and to understand their interaction with early AC tolerance. Studies should also collect longitudinal data on patient-reported outcomes and quality of life to balance efficacy with tolerability.

Mechanistic and translational studies

Embedded within clinical trials should be robust translational research components. This includes the collection of blood and tissue (both primary tumor and peritoneal fluid) for analysis of ctDNA dynamics, immune cell profiling, and tumor microenvironment changes to validate the biological mechanisms of early AC and identify new therapeutic targets[62].

Standardized endpoints and long-term follow-up

To overcome the biases seen in retrospective studies, endpoints such as peritoneal recurrence should be assessed by blinded, independent central review. Follow-up must extend beyond 3 years to capture late recurrence patterns and assess for potential long-term sequelae of early treatment.

CONCLUSION

The study by Lin et al[23] represents a crucial step in challenging the conventional paradigm of AC timing in advanced GC, demonstrating that ultra-early initiation is feasible under ERAS protocols and generating a provocative signal for reduced peritoneal recurrence. However, the path forward is not to simply advocate for early treatment for all patients. The study’s limitations—retrospective design, small sample size, and absence of molecular context—serve as a potent reminder of the complexities involved. The future of AC timing lies in precision. We must move from a “one-size-fits-all” approach to a biologically informed, dynamic, and individualized strategy. This precision timing framework, integrating tumor biology (molecular subtypes and ctDNA), patient recovery (ERAS metrics and nutritional status), and dynamic monitoring, offers the most promising pathway to improve outcomes. The clinical and scientific community now faces the imperative to design and execute large, well-powered, biomarker-driven prospective trials to validate ultra-early AC in high-risk subgroups and to establish the optimal, personalized treatment sequence for patients with resected GC. Only then can we translate the promise of feasibility into a tangible survival benefit.