Neoadjuvant Chemoradiotherapy for Locally Advanced Gastric Cancer: Where Are We at?

Simple Summary More than 50% of gastric cancer are at least locally advanced at presentation. For such patients, a multimodal approach rather than mere surgical resection leads to better long-term prognosis. Neoadjuvant chemoradiotherapy is one of the common treatment strategies for local advanced gastric cancer. Based on the experience and evidence from esophago-gastric cancers, the incorporation of systemic and locoregional therapy has shown superior disease control and reduced local recurrence. However, the optimal chemotherapy regimen, patient selection, technical consideration and potential biomarkers are still under investigation. Furthermore, the comparison of neoadjuvant chemoradiotherapy with neoadjuvant/perioperative chemotherapy is also an important issue to be answered. In the review article, we addressed the current available evidence to provide a comprehensive understanding and the use of neoadjuvant chemoradiotherapy for locally advanced gastric cancer. Future studies and ongoing trials will be necessary to determine the best candidate and the role of newer systemic and radiation therapies in such patients. NCRT is a feasible treatment option for LAGC, with the ability to achieve favorable disease control and enable higher radical resection rates over those afforded by perioperative chemotherapy or surgery alone. Large clinical trials examining the comparative efficacy of NCRT and NCT are underway. The discrepancy between the satisfactory pCR rates associated with NCRT and the nonsignificant association between NCRT and survival warrants further exploration. Furthermore, newer therapies such as immunotherapy and adaptive radiotherapy may be implemented in con-junction with NCRT, and the development of useful biomarkers may ultimately lead to the de-velopment of personalized treatments for LAGC. These research directions may lead to the dis-covery of the optimal approach to administering NCRT to patients with LAGC. They may also aid in the determination of the optimal candidates for undergoing NCRT. Abstract Locally advanced gastric cancer (LAGC) has a poor prognosis with surgical resection alone, and neoadjuvant treatment has been recommended to improve surgical and oncological outcomes. Although neoadjuvant chemotherapy has been established to be effective for LAGC, the role of neoadjuvant chemoradiotherapy (NCRT) remains under investigation. Clinical experience and research evidence on esophagogastric junction adenocarcinoma (e.g., cardia gastric cancers) indicate that the likelihood of achieving sustainable local control is higher through NCRT than through resection alone. Furthermore, NCRT also has an acceptable treatment-related toxicity and adverse event profile. In particular, it increases the likelihood of achieving an R0 resection and a pathological complete response (pCR). Moreover, NCRT results in higher overall and recurrence-free survival rates than surgery alone; however, evidence on the survival benefits of NCRT versus neoadjuvant chemotherapy (NCT) remains conflicting. For noncardia gastric cancer, the efficacy of NCRT has mostly been reported in retrospective studies, and several large clinical trials are ongoing. Consequently, NCRT might play a more essential role in unresectable LAGC, for which NCT alone may not be adequate to attain disease control. The continual improvements in systemic treatments, radiotherapy techniques, and emerging biomarkers can also lead to improved personalized therapy for NCRT. To elucidate the contributions of NCRT to gastric cancer treatment in the future, the efficacy, potential toxicity, predictive biomarkers, and clinical considerations for implementing NCRT in different types of LAGC were reviewed.


Introduction
One of the most common cancers, gastric cancer, constitutes a leading cause of cancerrelated death despite improvements in treatment and the widespread eradication of Helicobacter pylori [1][2][3]. The suboptimal prognosis of this disease is likely attributable to its aggressive biological behavior and to its frequently advanced stage at diagnosis (in more than 50% of cases) [4]. Although surgical resection provides the highest chance of recovery, it is usually insufficient or inapplicable for locally advanced gastric cancer (LAGC). A multimodal strategy includes systemic and local therapies that are based on the tumor characteristics [5,6]; it can induce disease control, facilitate complete resection, and improve survival outcomes [7]. This principle applies not only to initially resectable disease but also unresectable LAGC [8].
LAGC is typically defined as a tumor of the stomach or esophagogastric junction (EGJ); it is a type of histologically confirmed adenocarcinoma staged under the clinical tumor, node, and metastasis (TNM) staging system as cT3-cT4b, lymph node metastasis (N1-N3) without distant metastases (M0) [9]. In this context, tumors exhibiting mesenteric root invasion, para-aortic lymphadenopathy, or major vessel encasement are considered unresectable. For resectable disease, neoadjuvant chemotherapy (NCT) has demonstrated clear survival benefits over those of initial surgery, regardless of whether adjuvant chemotherapy was implemented [10][11][12]. Moreover, NCT might result in the downstaging of LAGC, facilitating subsequent resection [13][14][15]. Little information is available on the addition of radiotherapy, namely, neoadjuvant chemoradiotherapy (NCRT), to LAGC treatment programs.
A network meta-analysis concluded that combining radiotherapy and chemotherapy leads to more favorable local control relative to modality alone [16]. According to clinical trials on esophageal or EGJ adenocarcinomas, NCRT is associated with a significantly lower local failure rate and higher pathological complete response (pCR) and R0 resection rates in subsequent surgery [17][18][19]. Furthermore, NCRT results in a more satisfactory clinical response than NCT, suggesting its viability as a treatment modality. Prognostic data for LAGC are less abundant than those for esophageal and EGJ cancers. Notably, several clinical trials exploring the efficacy and safety of NCRT in LAGC are ongoing [20][21][22][23]. This paper presents a narrative review of studies on NCRT for LAGC, examining its efficacy, adverse effects, technical aspects, and perioperative and oncological outcomes. Furthermore, we discuss recommendations for the implementation of NCRT in this context.

Material and Methods
We searched the PubMed and Cochrane Library databases in February 2022 for Englishlanguage articles. The search strategy is described in Appendix A. In brief, the titles and abstracts of the publications retrieved from the aforementioned databases were screened. Subsequently, relevant articles were manually reviewed to identify other potentially eligible studies.

NCRT versus Surgery Alone
Multimodal treatment has been advocated for locally advanced EGJ and esophageal cancer because of the poor survival rate afforded by radical surgery alone [24][25][26]. Specifically, NCRT or perioperative chemotherapy is recommended for EGJ adenocarcinoma [27]. EGJ adenocarcinoma can be further classified as esophageal or gastric cancer, with a distinct staging system for each type of cancer under a staging system slightly different from the TNM staging system. In general, EGJ tumors are staged as gastric cancer if they extend more than 2 cm to the proximal stomach; otherwise, they are staged as esophageal cancer [28,29]. Under the Siewert classification, which is widely applied to the classification of EGJ cancers, type I and type II/III tumors are more appropriately staged as esophageal and gastric cancer, respectively [30]. However, gastric cardia cancers are frequently included with EGJ adenocarcinoma in clinical studies, and their management is largely the same.
The superior survival benefits conferred by NCRT over surgery alone for locally advanced esophageal cancer and EGJ cancer were first demonstrated in an Irish clinical trial in which 113 patients with esophageal adenocarcinoma were randomly assigned to receive either NCRT or surgery alone. The 3 year overall survival (OS) rate achieved through NCRT was significantly higher than that achieved through surgery (32% vs. 6%, p = 0.01) [31]. In another trial, CALGB 9781, in which patients with esophageal adenocarcinoma constituted the majority, NCRT also resulted in more favorable survival over surgery alone (median OS, 4.48 vs. 1.79 years, p = 0.02) [32]. Moreover, in the phase III CROSS trial [33], NCRT was associated with a higher R0 resection rate (92% vs. 69%, p < 0.001) and OS (hazard ratio [HR] 0.65, 95% confidence interval [CI] 0.49-0.87) than surgery alone. The rate of major adverse events associated with NCRT was acceptable (6% leukopenia and 5% anorexia), and inhospital mortality did not differ between the two groups. Furthermore, the survival benefits afforded by NCRT persisted over 10 years in the long-term follow-up [34]. Although both squamous cell cancer and adenocarcinoma were considered in the trial, 75% of the patients had adenocarcinoma, and NCRT led to survival benefits in both types of cancer.
Other trials have reported negative results for NCRT. Aside from two studies that were underpowered due to the low number of cases [35,36], the FFCD 9901 trial, which included patients with stage I and II esophageal cancer, found that NCRT did not provide any survival benefits over surgery alone. Instead, it reported a significantly higher postoperative mortality rate of 11.1% of NCRT versus 3.4% of NCT (p = 0.049) [37]. These discrepant findings may be explained by between-study differences in patient characteristics; only 29.2% of the patients had adenocarcinoma, and most tumors were located at the middlethird of the esophagus. Although subgroup analysis for stage I and II tumors was not performed, the present study indicated that NCRT should be considered with caution for earlier stage disease. On the other hand, meta-analyses have consistently indicated that NCRT confers greater survival benefits than surgery alone for locally advanced esophageal and EGJ adenocarcinoma [38][39][40], and that these benefits may be more pronounced in younger patients (patients aged ≤55 years) [39]. The clinical studies discussed thus far are summarized in Table 1.

NCRT versus NCT for EGJ and Gastric Cardia Cancers
Since the MAGIC trial reported that perioperative chemotherapy with ECF regimen (i.e., epirubicin, cisplatin, and fluorouracil) resulted in a significantly more favorable clinical response and significantly higher OS over surgery alone for distal esophageal and gastric cardia adenocarcinoma [10], researchers have devoted efforts to determining whether NCRT or NCT is more suitable for gastric cardia cancers. The German POET trial is the only randomized controlled Phase III trial designed for EGJ cancer that compares NCRT and NCT [19,41]. Patients undergoing NCRT had a higher rate of local recurrence-free survival (RFS; HR 0.37, 95% CI 0.16-0.85) as well as a higher rate of pCR (14.3% vs. 1.9%, p = 0.03) and a trend toward higher 5 year OS (39.5% vs. 24.4%, HR 0.65, 95% CI 0.42-1.01). Notably, the subgroup analysis suggested that patients with cardia cancers (Siewert type II) gained more benefits from NCRT relative to patients with Siewert type I cancers.
Conversely, the phase III NEO-AEGIS trial [42] and a Swedish trial [18] indicated that NCRT did not confer greater benefits in terms of OS and RFS than NCT, despite the association of NRT with higher pCR and R0 resection rates. Moreover, a meta-analysis suggested that NCRT is associated with higher postoperative mortality rates than is NCT (relative risk (RR) 1.58, 95% CI 1.00-2.49) [16]. In summary, evidence from locally advanced EGJ cancer indicates that NCRT is the modality of choice in terms of local control, although whether it affords greater survival benefits over NCT remains unclear. Until more evidence from clinical trials is presented, the implementation of NCRT in cases of gastric cardia cancer can be considered [19,42,43].
Recently, the results of the recent phase II/III FLOT4 trial [44] suggest a new standard for perioperative chemotherapy for EGJ cancers and LAGC. The perioperative FLOT regimen, which comprises fluorouracil plus leucovorin, oxaliplatin, and docetaxel, provided superior OS (median, 50 vs. 35 months, HR: 0.77, 95% CI 0.63-0.94) relative to the ECF or ECX (i.e., epirubicin, cisplatin, and capecitabine) regimens. Although numerous patients may benefit from perioperative FLOT, whether it can be a substitute for NCRT remains unclear [44], and a clearer answer may emerge after the completion of the ESOPEC trial, which directly compares the perioperative FLOT and CROSS regimens.

NCRT for Locally Advanced, Resectable Noncardia Gastric Cancer
Based on the experience of and evidence from research on EGJ and cardia cancers, the main advantage of NCRT is that it achieves a higher rate of local control to enable subsequent curative surgery. Compared with its use in EGJ and cardia cancers, the use of NCRT for noncardia gastric cancer is less validated due to the lack of phase III randomized controlled trials. Evidence from mostly uncontrolled studies [13,[45][46][47][48][49][50][51][52] indicates that NCRT led to R0 resection and pCR rates of approximately 70-80% and approximately 20-25%, respectively. A review of the performances of NCRT and other modalities is presented as follows.
A recent randomized controlled trial examined adjuvant XELOX chemotherapy administered to 60 patients with LAGC and compared the outcomes of adjuvant XELOX chemotherapy with and without NCRT [56]. NCRT resulted in a significantly higher RFS rate (60.0% vs. 33.3%, p = 0.019) and a significantly lower local recurrence rate (11.5% vs. 36.7%, p = 0.039) for up to 3 years, without an increase in perioperative complications (23.1% vs. 30.0%, p = 0.560). No significant difference in OS was observed (63.3% vs. 50.0%, p = 0.215). These findings, which are summarized in Table 2, indicate that NCRT is more effective than adjuvant treatments in achieving and maintaining local control. To determine whether long-term OS can be improved under NCRT, further investigations are warranted.

NCRT versus NCT for Resectable LAGC
NCT has demonstrated survival benefits over upfront surgery [10,57], and this approach is embraced under current treatment guidelines [8]. Nonrandomized studies have compared NCRT and NCT (Table 3) [46,51,52,54,58,59]. In general, NCRT is more likely to achieve a pCR and enable R0 resection than is NCT. However, the advantage of more favorable local control does not confer OS benefits. Trumbull et al. [59] observed greater survival benefits in terms of pCR under NCT than under NCRT in patients with LAGC. Another study found that NCRT was beneficial in patients with initial lymph node metastasis, as indicated by the association of a complete nodal response with significantly improved survival. Martin-Romano et al. reported that compared with patients receiving NCT, patients receiving NCRT had a higher likelihood of achieving a better local response (Becker Ia-b response, 58 vs. 32%, p = 0.001), exhibiting a grade D nodal regression (30% vs. 6%, p = 0.009), and attaining a favorable pathological response (23% vs. 3%; p = 0.019). However, no between-group difference in survival in patients with no cancerous lymph nodes at baseline was detected [51].  Reasons explaining why more favorable disease control does not necessarily translate to longer survival include insufficient follow-up durations, differences in chemotherapeutics, selection bias ascribable to a retrospective design, and potential toxicity attributable to radiotherapy. Furthermore, whether the NCRT outcomes vary with the location of the tumor remains unclear. A study based on the US National Cancer Database found that NCRT resulted in less favorable outcomes than perioperative chemotherapy (HR 1.10, 95% CI 1.00-1.20) in proximal gastric cancer. Corresponding data for distal gastric cancer are lacking [60]. Notably, NCRT remains recommended as an alternative to NCT in the guidelines of the US National Comprehensive Cancer Network. At least two large randomized controlled trials (i.e., TOPGEAR and CRITICS-II) evaluating the feasibility of NCRT for resectable LAGC are ongoing [20,61].

NCRT versus NCT for Unresectable LAGC
Less uncertainty surrounds the optimal approach for unresectable LAGC. In general, treatment should be administered to improve survival rates and quality of life and to palliate symptoms. For medically fit patients, systemic therapies are frequently recommended. Among various modalities, NCRT has been reported to result in a wide-ranging response rate (33-83%, Table 4) [62][63][64][65][66], and under the exclusion of metastatic disease, a clinical complete response was observed in 23-36% of patients [63,64]. In a phase II study including 29 patients, the initial NCRT response rate was 55%, and no clinical complete response was achieved. However, R0 resection was attained in 10 patients undergoing subsequent surgical resection, and pCR was achieved in 4 of those individuals [62]. GC: gastric cancer; pCR: pathological complete response; OS: overall survival; RFS: recurrence-free survival; RT: radiation therapy; EBRT: external beam radiation therapy; IMRT: intensity modulated radiation therapy; NCT: neoadjuvant chemotherapy; PCT: perioperative chemotherapy; DCF: doxorubicin, cisplatin, and 5-fluorouracil; FOLFOX: fluorouracil, leucovorin, and oxaliplatin; US: United States; HR: hazard ratio; CI: confidence interval.
An observational study was conducted to compare NCRT and NCT in unresectable and metastatic LAGC. Compared with NCT, NCRT resulted in significantly longer OS (median 14 months vs. 10 months, p = 0.011) and RFS (median 9 months vs. 8 months, p = 0.008). The occurrence of grade 3/4 toxicity did not differ significantly between the two groups [65]. A recent retrospective study determined that in patients with LAGC, compared with NCT, neoadjuvant chemotherapy was associated with a more favorable pathologic response, with no increase in serious postoperative complications [67]. Furthermore, an investigation based on the US National Cancer Database found that in cases of in unresectable LAGC, NCRT was associated with higher survival rates relative to chemotherapy alone. Only 30.8% of patients received NCRT in that study, suggesting the potential underutilization of this modality [66].
Taken together, the evidence indicates that NCRT is a viable option for patients with unresectable LAGC with favorable performance status. However, further randomized controlled trials are warranted to explore the optimal approach to treating this patient population.

Selection of Chemotherapy Regimen of NCRT for LAGC
The optimal chemotherapy regimen for NCRT has yet to established. The primary regimen used in clinical trials is doublet chemotherapy; the types of doublet chemotherapy include cisplatin plus 5-fluorouracil [32], capecitabine plus oxaliplatin [68], and the CROSS regimen, which comprises low-dose carboplatin plus paclitaxel [33] and is recommended by the latest NCCN guidelines [8]. Although a direct comparison of the aforementioned treatments has yet to be conducted, the CROSS regimen is suggested by various experts because of its ease of administration (single weekly dose) and favorable safe profile, especially for mucositis. Furthermore, a recent network meta-analysis of esophageal cancer also suggested that the CROSS regimen is preferred over a 5-fluorouracil-based regimen [69]. In addition, whether induction chemotherapy followed by chemoradiation can improve the efficacy of NCRT remains unclear [58,70], and this topic may be further clarified by the ongoing CRITICS-II trial [61].

Toxicity, Therapeutical Considerations, and Biomarkers of NCRT
NCRT is generally considered well tolerated, and overall complication rates have been noted to be comparable to those of NCT [46,65,67,71] and surgery alone [71][72][73]. However, a study based on a US national database suggested that NCRT is associated with significantly higher 30 day postoperative mortality rates (2.91% vs. 1.68% and 0.04%, p < 0.001) and significantly higher 90 day postoperative mortality rates (7.09% vs. 4.63% and 0.39%, p < 0.001) than perioperative chemotherapy or adjuvant chemoradiotherapy. In that study, NCRT was primarily administered to patients with proximal LAGC. This finding may partly explain why NCRT did not produce significant survival benefits despite providing more favorable local disease control in other studies [46,51,52,54,58,59]. However, this supposition necessitates validation through further large-scale clinical trials. In addition, Fujitani et al. suggested that an age of >60 years (RR: 11.3, 95% CI: 2.50-50.6) and a body mass index of ≥26 (RR: 4.08, 95% CI: 1.08-15.4) are significant risk factors for postoperative complications in patients who underwent NCRT following induction chemotherapy [74].
Radiotherapy, which is a crucial component of NCRT, has evolved as a medical treatment [75]. In clinical practice, radiotherapy is administered on the basis of tumor location and node metastasis status. For proximal gastric cancer, the field must cover the splenic hilum and paraesophageal lymph nodes; as for distal gastric cancer, coverage should encompass the porta hepatis, the distal superior mesenteric artery, and the duodenal stump [76]. The use of radiotherapy was evaluated because the earlier INT0116 trial suggested successful local control and a potential survival benefit with adjuvant chemoradiotherapy compared with surgery alone, even when old parallel-opposed anterior and posterior field arrangements were implemented [77]. Subsequently, a study demonstrated that the newer three-dimensional confocal radiotherapy can be safely administered with the more effective ECF chemotherapy regimen [78], and it has become the backbone of the TOPGEAR trial [20]. Notably, the role of adjuvant chemoradiotherapy has decreased because of its lack of advantages versus perioperative chemotherapy [21,79], and ongoing studies are focusing on the role of NCRT.
Modern radiotherapy includes dynamic planning techniques (e.g., image-guided radiotherapy, adaptive radiotherapy [80], and four-dimensional computed tomography [81]) because of the mobility of the stomach. Furthermore, intensity-modulated radiotherapy (IMRT) may reduce extra-organ damage to healthy tissues adjacent to the treatment site better than conventional three-dimensional conformal radiotherapy [82,83]. The combination of newer planning and therapeutic modalities can minimize the radiation dose and associated toxicity. Adaptive radiotherapy, which uses deformation models to adapt to individual gastric shapes, has been validated, and a population-based model is available [84]. The progress achieved can improve the safety and effectiveness of the treatment in a personalized manner.
Several biomarkers have been associated with the outcomes of NCRT-treated LAGC, including overexpressed ERCC1 and ERCC2 [4], LAG-3 [85], microRNAs (miRs) 338-3p and miR-142-3p [86], T-cell density, and 5-fluorouracil-related enzymes [87,88]. Before these biomarkers are applied to clinical practice, further validation of these findings is required. Notably, the presence of signet ring cell carcinoma and a higher histologic grade may be associated with chemoradiation resistance and poorer prognosis [89]. Shared decision making and participation in clinical trials should thus be considered for such patients whenever possible.

Conclusions and Future Perspectives
NCRT is a feasible treatment option for LAGC, with the ability to achieve favorable disease control and enable higher radical resection rates over those afforded by perioperative chemotherapy or surgery alone. Large clinical trials examining the comparative efficacy of NCRT and NCT are underway. The discrepancy between the satisfactory pCR rates associated with NCRT and the nonsignificant association between NCRT and survival warrants further exploration. Furthermore, newer therapies such as immunotherapy and adaptive radiotherapy may be implemented in conjunction with NCRT, and the development of useful biomarkers may ultimately lead to the development of personalized treatments for LAGC. These research directions may lead to the discovery of the optimal approach to administering NCRT to patients with LAGC. They may also aid in the determination of the optimal candidates for undergoing NCRT.  Data Availability Statement: All data were directly extracted or appropriately calculated based on the cited articles.

Conflicts of Interest:
The authors declared no potential conflict of interest with respect to the research, authorship, and/or publication of this article.