Impact of prior cancer history on survival in patients with adenocarcinoma of esophagogastric junction: a retrospective cohort study using SEER database

Introduction

The exclusion of patients with prior malignancies from clinical trials represents a significant challenge in oncology research, particularly impacting the generalizability of findings to real-world populations (1). While this practice aims to minimize confounding in survival analyses, it may systematically underrepresent important patient subgroups, including older adults and those with complex medical histories. Specifically, patients with prior cancers often exhibit distinct biological and therapeutic trajectories due to cumulative treatment-related damage (e.g., radiation fibrosis or chemotherapy-induced organ dysfunction), immune modulation from prior antitumor therapies, and underlying genetic susceptibility (1-7). In the context of adenocarcinoma of esophagogastric junction (AEG), thoracic radiation sequelae may impair physiological reserve, while prior immunotherapies can alter tumor microenvironment responses (3,8,9). The junction’s anatomical vulnerability further shape disease behavior. These factors collectively influence AEG treatment tolerance and progression patterns, warranting population-specific investigation.

In AEG, a histologically and epidemiologically distinct subtype with rising incidence, the prognostic implications of prior malignancies remain poorly characterized (10-16). AEG’s unique anatomical position at the gastroesophageal junction confers distinct biological behaviors that necessitate separate analysis from gastric or esophageal cancers generally (17). This discordance underscores the need for AEG’s unique features—including its junctional location altering lymphatic drainage patterns, molecular heterogeneity (e.g., higher HER2 positivity compared to distal gastric cancers), and requirement for multidisciplinary management balancing esophageal and gastric paradigms—mandate distinct prognostic considerations. These distinctions necessitate separate analysis rather than extrapolation from other gastrointestinal malignancies.

Established prognostic factors for AEG include Siewert classification (types I–III), HER2 status, response to neoadjuvant therapy, and molecular subtypes, all of which significantly influence treatment decisions and outcomes (10,18-23). Several well-established prognostic factors demonstrate complex interactions with prior cancer history. Siewert classification stratifies anatomical localization and predicts lymphatic spread and treatment responses, though prior abdominal surgeries or radiotherapy may necessitate modified surgical strategies (10,17,24). HER2 overexpression (15–30% of AEG cases), a key determinant of targeted therapy efficacy as validated in the ToGA trial, may be dynamically altered by prior therapies through epigenetic modulation or clonal selection (3,25,26). Response to neoadjuvant therapy (e.g., CROSS or FLOT regimens) serves as another critical prognostic indicator, while prior cancer treatments may compromise tolerance to standard neoadjuvant regimens due to cumulative toxicities, potentially altering this prognostic relationship (27).

Existing studies on gastrointestinal cancers demonstrate heterogeneous outcomes: while prior cancers of the lung or colon are associated with worse survival in colorectal cohorts, no such association exists for esophageal or breast primaries (2,28,29). Conflicting evidence persists: Laccetti et al. performed a retrospective study demonstrating that a history of prior cancer did not adversely affect the clinical outcomes of patients with advanced lung cancer, regardless of the stage or type of prior cancer (30). However, it was also reported that overall survival rates were significantly lower in gastric cancer patients with prior cancer compared to those without (31,32). This discordance underscores the need for site-specific analyses (33). Furthermore, current trial designs often conflate all prior malignancies into a monolithic exclusion category, disregarding critical variables such as primary tumor site, stage, latency period, and treatment modality—factors demonstrated to modulate survival in other cancers (1,2).

This study addresses two critical gaps in understanding survival and trial participation among AEG patients with prior malignancies. We hypothesize two potential mechanistic pathways: (I) a “cumulative toxicity” pathway where prior cancer treatments (e.g., anthracycline-induced cardiotoxicity, radiation fibrosis) compromise physiological reserve and tolerance to subsequent AEG therapies; and (II) a “surveillance effect” pathway where intensive follow-up for prior malignancies leads to earlier AEG detection and stage migration. First, it quantifies how characteristics of prior cancers—including primary site, stage, and latency—shape AEG-specific survival, particularly in the setting of modern multimodal therapy where treatment interactions are complex. Second, it evaluates whether the routine exclusion of AEG patients with prior cancers systematically skews clinical trial populations toward biologically favorable subgroups, thereby compromising the generalizability of trial results to real-world patient cohorts. By leveraging population-level data and multivariable competing risk regression models, this population-based study evaluates the prevalence of prior cancer and assesses its impact on survival in 10,895 AEG patients, aiming to inform evidence-based adjustments to trial eligibility thresholds and support personalized treatment strategies for this understudied population. We present this article in accordance with the STROBE reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2046/rc).

Methods

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. SEER*Stat software was utilized to extract data from the Surveillance, Epidemiology, and End Results (SEER) Research Data (November 2023 Submission) covering the period from 1975 to 2021 (Version 8.4.3). The study cohort included patients diagnosed with AEG between 1975 and 2019, identified using ICD-O-3 topography codes 160 and “CS site-specific factor 25”, with histology codes 8140-8147, 8210-8221, 8255, 8260-8263, 8310, 8480-8481, 8570-8574, 8568, and 8490. Detailed information can be found on the official website (https://seer.cancer.gov/). A total of 10,895 eligible patients were included after applying exclusion criteria (Figure 1).

Figure 1 Study flowchart. SEER, Surveillance, Epidemiology, and End Results.

Patients

Data on patient demographics, histological features, and follow-up were collected. Figure 1 illustrates the established inclusion criteria: (I) demographic and histopathological data were completed; (II) the histopathological diagnosis is AEG, with the anatomical position code in the tumor disease system (ICD-O-3) being 160, and “CS site-specific factor 25”; the histological codes include 8140-8147, 8210-8221, 8255, 8260-8263, 8310, 8480-8481, 8570-8574, 8568, and 8490; (III) follow-up completed (34-37). Exclusion criteria: (I) patients with a history of multiple primary cancers (=3,587); (II) patients with a survival duration of less than 1 month.

The history of previous cancers was determined using the SEER number sequence recode, which records the sequence of all tumours during the patient’s their lives (31,32). Patients with sequence number “00” indicated a single primary cancer (AEG only), while sequence numbers “01” or higher identified those with prior malignancies. For patients with multiple neoplasms, we included only those with one prior cancer preceding the index AEG diagnosis to ensure temporal clarity. The timing interval between prior cancer and AEG diagnosis was calculated using SEER diagnosis dates, with validation through consistency checks between sequence numbers and corresponding diagnosis dates. This approach aligns with SEER’s coding rules and prior validation studies (31,32,38).

To address heterogeneity, prior cancers were classified according to: (I) primary site (prostate, breast, lung/bronchus, colorectal, urinary bladder, etc.); (II) SEER summary stage (localized, regional, distant); and (III) time interval from AEG diagnosis (<1, 1–2, >2–3, >3–4, >4–5, >5 years); (IV) period of diagnosis time (1975–2003, 2004–2019) based on established clinical classifications. We also calculated the timing of the prior cancers using the SEER diagnosis dates of the index cancer and the most recent of any prior cancers (38). To control for potential confounding effects, we performed multivariable Cox proportional hazards regression analyses that included primary site and SEER stage of prior cancers as categorical variables. Primary site was classified into major categories (prostate, breast, lung/bronchus, colorectal, urinary bladder, other), while SEER stage was categorized as localized, regional, or distant. These variables were adjusted for in all survival models to isolate the independent effects of prior cancer characteristics on AEG outcomes.

The classification of treatment modalities (chemotherapy and radiation therapy) in SEER requires careful interpretation due to inherent database constraints. SEER data have documented limitations in capturing chemotherapy administration, with the variable coded as a binary classification: “Yes” for definitively documented treatment versus “No/Unknown” for cases where treatment status is ambiguous or unrecorded. This combined category reflects the operational reality of SEER coding, where “Unknown” cannot be separated from “No” due to systemic data capture issues.

Patient demographic data included age, race, gender, and year of diagnosis. Histopathological characteristics included histological type, tumour grade, tumor size, SEER stage, and TNM stage. According to the SEER staging system, disease extension was categorized as localized, regional and distant. Localised disease denotes cancer that is restricted to the esophagogastric junction. Regional disease includes cancer characterised by direct tumour extension to neighbouring structures or the involvement of regional lymph nodes. Distant disease encompasses any type of metastatic cancer. The 6th AJCC TNM staging system was employed to consistently apply the TNM stage to AEG, irrespective of the tumor epicenter at the esophagogastric junction, facilitating a sensitive analysis of tumor stage. The treatment regimen included radiation treatment, chemotherapy, and tumour excision. SEER data have documented limitations in capturing chemotherapy administration, with the variable coded as a binary classification: “Yes” for definitively documented treatment versus “No/Unknown” for cases where treatment status is ambiguous or unrecorded. Therefore, we conducted stratified analyses comparing outcomes between patients with definitive “Yes” chemotherapy versus the composite “No/Unknown” group, while acknowledging potential misclassification. Follow-up data encompassed overall survival (OS) and cancer-specific survival (CSS). OS indicates the time span from diagnosis to death from any cause, whereas CSS refers to the period from diagnosis to death specifically due to AEG.

Statistical analysis

Patients with adenocarcinoma of the esophagogastric junction were stratified into two cohorts based on prior cancer history (with/without prior malignancy). Baseline tumor characteristics (SEER stage, TNM stage, histological grade) and treatment modalities (surgery, chemotherapy) for both primary and secondary cancers were systematically documented. Statistical analysis was conducted using IBM SPSS Inc.’s SPSS version 25.0 and R software version 4.1.1. Continuous variables were analysed between groups using Student’s t-tests, whereas categorical variables were evaluated with Fisher’s exact test or Pearson’s Chi-squared test. Survival across different groups was evaluated using the Kaplan-Meier method and log-rank tests. To minimize confounding and maximize statistical power in this large cohort, the univariate and multivariable Cox proportional hazards regression models were employed as the primary method to examine the influence of prior cancer history on OS and CSS. Variables with statistical significance (P<0.05) in univariate analysis or clinical relevance were included in the multivariate models. The Fine and Gray competing risk regression model was used to account for competing risks in CSS analysis.

Propensity score matching (PSM) was applied as a sensitivity analysis to validate the robustness of the primary findings. PSM was conducted using a 1:1 nearest-neighbor matching algorithm without replacement, with a caliper width of 0.05 standard deviations of the propensity score logit, to balance covariates between groups including age, race, tumor grade, year of diagnosis, histological type, treatment (chemotherapy, radiation, surgery), SEER stage. This approach ensured that the primary analysis retained the full cohort’s representativeness, while PSM provided supplementary evidence of consistency. A two-sided P value of less than 0.05 was considered statistically significant unless otherwise specified.

Results

Baseline characteristics

A total of 10,895 patients diagnosed with AEG between 1975 and 2019 were eligible for inclusion in this study. Among these, 1,806 patients (16.58%) were diagnosed with prior cancer, with a median age of 73 years [interquartile range (IQR): 67–83 years] at the time of AEG diagnosis. The baseline characteristics are summarized based on Table 1 (intra-patient comparison of features at prior cancer diagnosis versus AEG diagnosis) and Table 2 (inter-group comparison between patients with and without prior cancer, before and after propensity score matching).

As shown in Table 1, it was found that patients with a history of cancer were more likely to present with early stage at the diagnosis of their first primary cancer (staged with localized SEER stage: 38.90% vs. 23.30%; P<0.001). Additionally, these patients were less likely to receive radiation (30.30% vs. 37.60%; P<0.001) and chemotherapy (17.60% vs. 46.50%; P<0.001) as treatment for their first primary cancer. The median interval between prior cancer and AEG diagnoses was 6.00 years (IQR, 2.00–13.00 years).

Table 2 indicates that compared with patients without prior cancer, those with prior cancer at the time of their AEG diagnosis were more likely to be older and have a lower proportion of G1/G2 tumor grade (G1/G2: 30.90% vs. 32.10%, P=0.001). Patients with a history of cancer exhibited a greater proportion of early-stage tumors (localized stage: 23.30% vs. 17.50%, P<0.001) and a higher percentage of tumors with diameters less than 5 cm compared to those without prior cancer (18.20% vs. 15.90%, P=0.03). Besides, patients with a history of cancer exhibited a higher likelihood of declining chemotherapy (receipt of chemotherapy: 46.50% vs. 57.50%, P<0.001) and radiation therapy (receipt of radiation: 37.60% vs. 42.50%, P<0.001). This may indicate age- and stage-related effects, as patients without prior cancer had a greater proportion of advanced tumors and were more commonly younger, while those with a history of cancer were predominantly older and demonstrated increased intolerance to systemic therapy. After matching, all variables—including age—showed no significant differences between groups (all P>0.05), confirming successful balancing of covariates. Malignancies of the prostate (28.07%), colon and cecum (11.13%), breast (8.42%), and urinary bladder (8.31%) were the most prevalent prior cancer types (as shown in Figure 2).

Figure 2 Distributions of prior cancer types in patients with AEG. AEG, adenocarcinoma of esophagogastric junction.

Survival analysis and multivariable analysis

Survival analysis was performed using the Kaplan-Meier method, univariate/multivariate Cox regression, and competing risk models (Fine and Gray). The median CSS and OS for patients were 15 months and 13 months, respectively. Deaths related to AEG were observed in 7,784 patients (71.40%), whereas 9,146 patients (84.00%) succumbed to all causes combined. Figure 3 illustrates the impact of prior cancer on OS and CSS via Kaplan-Meier curves. Before PSM: Patients with prior cancer had significantly worse OS than those without (log-rank P<0.001; Figure 3A), but no difference in CSS (log-rank P=0.17; Figure 3C). After PSM, baseline characteristics were balanced (Table 2), and the association persisted: prior cancer remained linked to inferior OS (log-rank P=0.01; Figure 3B) but not CSS (log-rank P=0.29; Figure 3D).

Figure 3 Kaplan-Meier survival curves of prior cancer impact on the OS and CSS in AEG patients with or without prior cancer. (A) The OS analysis before PSM; (B) the OS analysis after PSM; (C) the CSS analysis before PSM; (D) the CSS analysis after PSM. AEG, adenocarcinoma of esophagogastric junction; CSS, cancer-specific survival; OS, overall survival; PSM, propensity score matching.

Then, univariate and multivariate Cox regression were used to evaluate the impact of prior cancer on OS (Table 3) and CSS (Table 4) in AEG patients and confirm the consistency of the result of PSM above. A single composite variable could obscure important differences inherent in the simple presence or absence of a prior cancer, its anatomic extent (SEER stage) at diagnosis, and its underlying biology (cancer type). To comprehensively evaluate the impact of a prior cancer history on OS while accounting for its clinical heterogeneity, we constructed three separate multivariable Cox proportional hazards models. Each model adjusted for the same set of covariates but differed solely in the variable used to characterize the prior malignancy: Model a used the binary variable prior (any prior cancer history); Model b used the categorical variable SEER stage of prior cancer; and Model c used the specific Prior cancer types.

After adjusting for age, race, year of diagnosis, tumor grade, SEER stage, and treatment modalities (surgery, radiation, chemotherapy), the presence of any prior cancer history was significantly associated with inferior OS compared to having no prior cancer (HR =1.16, 95% CI: 1.10–1.23, P<0.001) shown in Table 3. When the prior cancer was characterized by its SEER stage at diagnosis, using “None” as the reference, a significant association with OS was observed for prior cancers diagnosed at the regional stage (HR =1.22, 95% CI: 1.07–1.40, P=0.002). Analysis based on the specific type of the prior cancer revealed heterogeneous associations with OS. Using “No prior cancer” as the reference, notably, specific cancer types including breast (HR =1.00, 95% CI: 0.84–1.20, P=0.97), and lymphoma (HR =1.17, 95% CI: 0.89–1.54, P=0.27) showed neutral on overall survival.

However, using similar multivariate Cox regression analysis for CSS shown in Table 4, it was found that prior cancer (HR =1.03, 95% CI: 0.96–1.10, P=0.43) and its subvariables (SEER stage of prior cancer, SEER stage of prior cancer) did not consistently demonstrate prognostic significance in the multivariable models. Notably, competing risk analysis (Table 5) demonstrated that a prior cancer history independently predicted improved AEG cancer-specific survival (HR = 0.83, 95% CI: 0.77–0.89, P<0.001). Table 5 further elucidated the heterogeneity of this association across subgroups defined by prior cancer timing, stage, and histological type. Concordant with Kaplan-Meier survival analysis, multivariate Cox regression (Table 5) indicated that, compared with patients without prior cancer, those with prior cancer of varying timing, stages, or histological types did not exhibit inferior overall survival (as reflected by adjusted hazard ratios close to 1). However, notable exceptions emerged: patients with prior non-localized cancer, or prior cancer of the lung and bronchus, colon and cecum, or urinary bladder, as well as those with a prior cancer diagnosis interval >5 years or <2–3 years, showed compromised overall survival relative to patients without prior cancer. Similar results were also found when the prior cancer was characterized by its SEER stage at diagnosis as shown in Table 6.

Subgroup analysis: stratification by timing of prior cancer interval

Subgroup analyses further clarified differential impacts of prior cancer characteristics such as timing, SEER stage of previous cancer, and type of previous cancer.

When stratified by interval between prior cancer and AEG diagnosis, KM analysis (Figure 4) visually depicts the differential impact of the interval between prior cancer and AEG diagnoses on survival. Curves show that OS is significantly impaired when the interval is 2–3 or >5 years, whereas CSS remains largely unaffected except for intervals <1 year, which exhibits a transient protective effect. Table 5 validation quantifies these observations: Intervals of 2–3 years (HR =1.30, 95% CI: 1.08–1.57, P=0.006) and >5 years (HR =1.25, 95% CI: 1.16–1.35, P<0.001) are independent risk factors for OS. Only intervals <1 year show significant CSS improvement (HR =0.58, 95% CI: 0.49-0.68, P<0.001), aligning with the KM curves where short-interval patients have better cancer-specific outcomes. Similar results were also found when the prior cancer was characterized by its SEER stage at diagnosis as shown in Table 6.

Figure 4 Kaplan-Meier survival curves of prior cancer impact on the OS and CSS stratified by timing of prior cancer in patients with AEG. (A) The OS analysis with time interval less than 1 year; (B) the OS analysis with time interval between 1 and 2 years; (C) the OS analysis with time interval between 2 and 3 years; (D) the CSS analysis with time interval less than 1 year; (E) the CSS analysis with time interval between 1 and 2 years; (F) the CSS analysis with time interval between 2 and 3 years; (G) the OS analysis with time interval between 3 and 4 years; (H) the OS analysis with time interval between 4 and 5 years; (I) the OS analysis with time interval longer than 5 years; (J) the CSS analysis with time interval between 3 and 4 years; (K) the CSS analysis with time interval between 4 and 5 years; (L) the CSS analysis with time interval longer than 5 years. AEG, adenocarcinoma of esophagogastric junction; CSS, cancer-specific survival; OS, overall survival.

Subgroup analysis: stratification by SEER stage of prior cancer

When stratified by SEER stage of prior cancer (localized/regional/distant), KM analysis (Figure 5) demonstrates that patients with non-localized (regional/distant) prior cancer have markedly worse OS compared to those without prior cancer, whereas those with localized prior cancer show comparable or even improved CSS. Tables 4,6 validation confirm the stage-dependent effect. Regional stages confer higher OS risk (HR =1.22, 95% CI: 1.07–1.40, P=0.002), while localized prior cancer associates with CSS benefit (HR =0.69, 95% CI: 0.57–0.84, P=0.02) in competing risk analysis, consistent with KM curves where these patients exhibit longer CSS. Similar results were also found when the prior cancer was characterized by its SEER stage at diagnosis as shown in Table 6.

Figure 5 Kaplan-Meier survival curves of prior cancer impact on the OS and CSS stratified by stage of prior cancer in patients with AEG. (A) The OS analysis with prior cancer at localized stage; (B) the OS analysis with prior cancer at regional stage; (C) the OS analysis with prior cancer at distant stage; (D) the CSS analysis with prior cancer at localized stage; (E) the CSS analysis with prior cancer at regional stage; (F) the CSS analysis with prior cancer at distant stage. AEG, adenocarcinoma of esophagogastric junction; CSS, cancer-specific survival; OS, overall survival.

Subgroup analysis: stratification by type of prior cancer

KM analysis (Figure 6) reveals that prior cancers originating from lung/bronchus, colon/cecum, and urinary bladder significantly reduce OS, while other types (e.g., breast) have minimal OS impact or even CSS advantages. Table 5 validation provides hazard ratios for key cancer types. Significant OS detriments: for lung/bronchus (HR =1.42, 95% CI: 1.14–1.76, P=0.001), colon/cecum (HR =1.25, 95% CI: 1.07–1.46, P=0.004), and leukemia (HR =1.70, 95% CI: 1.13–2.57; P=0.01). However, CSS benefits were observed for lung/bronchus (HR =0.68, 95% CI: 0.48–0.95, P=0.03), colon/cecum (HR =0.67, 95% CI: 0.53–0.84, P<0.001), and prostate (HR =0.87, 95% CI: 0.76–0.99, P=0.043). Similar results were also found when the prior cancer was characterized by its SEER stage at diagnosis as shown in Table 6.

Figure 6 Kaplan-Meier survival curves of prior cancer impact on the OS and CSS stratified by different types of prior cancer in patients with AEG. (A) The impact of prior prostate cancer on OS; (B) the impact of prior breast cancer on OS; (C) the impact of prior colon or cecum cancer on OS; (D) the impact of prior lung and bronchus cancer on OS; (E) the impact of prior prostate cancer on CSS; (F) the impact of prior breast cancer on CSS; (G) the impact of prior colon or cecum cancer on CSS; (H) the impact of prior lung and bronchus cancer on CSS; (I) the impact of prior kidney and renal pelvis on OS; (J) the impact of prior melanoma of the skin on OS; (K) the impact of prior corpus uteri on OS; (L) the impact of prior lymphoma on OS; (M) the impact of prior kidney and renal pelvis on CSS; (N) the impact of prior melanoma of the skin on CSS; (O) the impact of prior corpus uteri on CSS; (P) the impact of prior lymphoma on CSS. AEG, adenocarcinoma of esophagogastric junction; CSS, cancer-specific survival; OS, overall survival.

Subgroup analysis: stratification by year of diagnosis

KM analysis stratified by year of diagnosis (Table 5) demonstrates that patients with a prior cancer diagnosed between 2004–2019 exhibit significantly worse OS compared to those without prior cancer, regardless of diagnosis year. Specifically, patients with prior cancer diagnosed in 1975–2003 had HR of 1.18 (95% CI: 1.07–1.29, P=0.001) for OS, while those diagnosed in 2004–2019 showed a slightly lower but still significant OS risk (HR =1.13, 95% CI: 1.05–1.21, P=0.001). For AEG-specific survival, prior cancer was associated with better outcomes in both periods: (HR =0.79, 95% CI: 0.69–0.90, P=0.001) for 1975–2003 and (HR =0.82, 95% CI: 0.75–0.90, P<0.001) for 2004–2019. These findings suggest that the adverse impact of prior cancer on survival persists across diagnosis eras, though the magnitude of risk may diminish slightly in more recent years. Similar results were also found when the prior cancer was characterized by its SEER stage at diagnosis as shown in Table 6.

Subgroup analysis: stratification by receipt of chemotherapy

Analysis stratified by chemotherapy receipt revealed a differential effect of prior cancer history on survival outcomes between treatment subgroups. Among patients who received chemotherapy, prior cancer history was associated with significantly improved CSS (HR =0.76, 95% CI: 0.69–0.84, P<0.001) compared to patients without prior cancer. However, no significant association was observed for OS (HR =0.96, 95% CI: 0.88–1.04, P=0.29).

Among patients who did not receive or had unknown chemotherapy status, prior cancer history showed a more modest protective effect on CSS (HR =0.87, 95% CI: 0.78–0.97, P=0.02) but was associated with significantly worse OS (adjusted HR =1.36, 95% CI: 1.25–1.47, P<0.001) compared to those without prior cancer. Similar results were also found when the prior cancer was characterized by its SEER stage at diagnosis as shown in Table 6.

These results suggest that chemotherapy may modify the effect of prior cancer history, particularly enhancing the CSS benefit observed in patients with prior malignancies. The differential effects between OS and CSS outcomes highlight the complex interplay between prior cancer history, treatment status, and cause-specific mortality in AEG patients.

Discussion

This large-scale population-based cohort study addresses critical gaps in understanding survival dynamics among AEG patients with prior malignancies. Unlike prior studies conflating gastric and esophagogastric junction adenocarcinomas, our analysis elucidates distinct prognostic interactions between prior malignancies and AEG biology. Furthermore, this study represents one of the largest investigations involving over 10,000 AEG patients and utilizing a national database to examine the clinical characteristics, treatment, and survival outcomes of AEG patients with a history of cancer. The results may influence the treatment therapies offered by physicians to these patients and the criteria for participation in related clinical trials.

Discrepancy between clinical practice patterns and survival outcomes

The study reveals that 16.58% of AEG patients had prior malignancies, a prevalence comparable to gastric (18.80%) and lung cancers (14.70%) (30,31). Despite presenting at an older age, earlier SEER stage, and lower tumor grade (suggesting enhanced surveillance), these patients exhibited reduced uptake of multidisciplinary therapies (chemotherapy: 46.50% vs. 57.50%). This paradox highlights that prior cancer treatments (e.g., pelvic radiotherapy) may constrain tolerance to intensive regimens like neoadjuvant FLOT or immunotherapy. For instance, bladder cancer patients frequently exposed to radiotherapy may develop myelosuppression or gastrointestinal toxicity, limiting subsequent chemotherapy feasibility. Additionally, clinical caution driven by toxicity concerns may overshadow biology-based risk assessments.

Biological heterogeneity and compensatory mechanisms

Prior cancer was independently associated with worse overall survival (HR =1.16, 95% CI: 1.09–1.23, P<0.001) but not cancer-specific survival (HR =1.03, 95% CI: 0.96–1.10, P=0.43). However, Subgroup analyses revealed several patient populations where prior malignancy demonstrated no adverse prognostic impact. Patients with prior cancer diagnosed within 1 year showed comparable overall survival (HR =1.03, 95% CI: 0.91–1.15, P=0.67) and significantly better cancer-specific survival (HR =0.58, 95% CI: 0.49–0.68, P<0.001) in competing risk analysis. Those with localized-stage prior cancers exhibited no significant difference in cancer-specific survival (HR =0.91, 95% CI: 0.82–1.02, P=0.12) in competing risk analysis. Notably, specific cancer types including breast (HR =1.00, 95% CI: 0.84–1.20, P=0.97), and lymphoma (HR =1.17, 95% CI: 0.89–1.54, P=0.27) showed neutral effects on overall survival.

Our study design allows us to evaluate two non-mutually exclusive mechanisms that may explain the observed survival patterns: the cumulative toxicity hypothesis posits that prior cancer treatments may compromise physiological reserve and treatment tolerance for subsequent AEG therapy. Several findings support this mechanism: (I) patients with prior cancer received significantly less chemotherapy (46.5% vs. 57.5%, P<0.001) and radiation therapy (37.6% vs. 42.5%, P<0.001), suggesting potential treatment-limiting toxicities; (II) the differential impact on overall survival (worse) versus cancer-specific survival (similar) suggests non-cancer mortality drivers possibly related to treatment sequelae. The Surveillance Effect Hypothesis suggests enhanced detection capabilities through ongoing cancer follow-up may lead to stage migration. Supporting evidence includes: (I) prior cancer patients presented with earlier-stage AEG (localized stage: 23.30% vs. 17.50%, P<0.001) and smaller tumors (<5 cm: 18.20% vs. 15.90%, P=0.03); (II) the comparable cancer-specific survival despite reduced treatment intensity may indicate genuinely less aggressive disease at diagnosis; (III) the predominance of screen-detectable prior cancers (prostate 28.1%, breast 8.4%) in our cohort supports ongoing surveillance patterns. Our data suggest these mechanisms likely operate concurrently, with their relative importance varying by prior cancer type and timing. Future studies with detailed treatment records and serial imaging data are needed to quantify their individual contributions.

These findings challenge the default exclusion of all prior cancer patients from trials. Instead, they support individualized eligibility assessments based on specific prior treatments, time intervals, and residual organ function. This approach aligns with the NCI’s Precision Medicine Initiatives advocating for basket trials incorporating molecular profiling and comorbidity indices. Future studies should prospectively capture detailed treatment toxicity data and surveillance intensity metrics to better distinguish between these mechanistic pathways.

This retrospective analysis is limited by potential confounding factors (e.g., unmeasured comorbidities) and reliance on administrative databases. Prospective validation in registries like SEER-Medicare or international cohorts is warranted. Additionally, mechanistic studies exploring tumor-immune interactions or clonal evolution between prior and subsequent malignancies could elucidate the biological underpinnings of our findings. Furthermore, the classification of treatment modalities (chemotherapy and radiation therapy) in SEER requires careful interpretation due to inherent database constraints. SEER data have documented limitations in capturing chemotherapy administration, with the variable coded as a binary classification: “Yes” for definitively documented treatment versus “No/Unknown” for cases where treatment status is ambiguous or unrecorded. To address this limitation, we implemented several methodological adjustments: Sensitivity Analyses: We conducted stratified analyses comparing outcomes between patients with definitive “Yes” chemotherapy versus the composite “No/Unknown” group. Our results suggest that chemotherapy may modify the effect of prior cancer history, particularly enhancing the CSS benefit observed in patients with prior malignancies.

Conclusions

While prior cancer history adversely impacts overall survival in AEG patients, specific subgroups—particularly those with localized-stage or select cancer types—exhibit comparable outcomes. These findings suggest that current exclusion criteria may be overly restrictive, and support refining trial eligibility to include well-selected prior-cancer patients.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2046/rc

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2046/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2025-2046/coif). The authors have no conflicts of interest to declare.

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. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Patel RR, Parisi R, Verma V, et al. Association between Prior Malignancy Exclusion Criteria and Age Disparities in Cancer Clinical Trials. Cancers (Basel) 2022;14:1048. [Crossref] [PubMed]
2. Al-Husseini MJ, Saad AM, Mohamed HH, et al. Impact of prior malignancies on outcome of colorectal cancer; revisiting clinical trial eligibility criteria. BMC Cancer 2019;19:863. [Crossref] [PubMed]
3. Zeng M, Lin A, Jiang A, et al. Decoding the mechanisms behind second primary cancers. J Transl Med 2025;23:115. [Crossref] [PubMed]
4. Travis LB, Rabkin CS, Brown LM, et al. Cancer survivorship--genetic susceptibility and second primary cancers: research strategies and recommendations. J Natl Cancer Inst 2006;98:15-25. [Crossref] [PubMed]
5. Najafi M, Salajegheh A, Rezaeyan A. Bystander Effect and Second Primary Cancers following Radiotherapy: What are its Significances? J Med Phys 2017;42:55-6. [Crossref] [PubMed]
6. Zamarron BF, Chen W. Dual roles of immune cells and their factors in cancer development and progression. Int J Biol Sci 2011;7:651-8. [Crossref] [PubMed]
7. Vogt A, Schmid S, Heinimann K, et al. Multiple primary tumours: challenges and approaches, a review. ESMO Open 2017;2:e000172. [Crossref] [PubMed]
8. de Groot C, Beukema JC, Langendijk JA, et al. Radiation-Induced Myocardial Fibrosis in Long-Term Esophageal Cancer Survivors. Int J Radiat Oncol Biol Phys 2021;110:1013-21. [Crossref] [PubMed]
9. Pazo-Cid R, Pons PG. Integrating immunotherapy in the treatment of resectable gastric cancer: Are we on the right track? Med 2025;6:100543. [Crossref] [PubMed]
10. Hasegawa S, Yoshikawa T. Adenocarcinoma of the esophagogastric junction: incidence, characteristics, and treatment strategies. Gastric Cancer 2010;13:63-73. [Crossref] [PubMed]
11. Trivers KF, Sabatino SA, Stewart SL. Trends in esophageal cancer incidence by histology, United States, 1998-2003. Int J Cancer 2008;123:1422-8. [Crossref] [PubMed]
12. Blot WJ, Devesa SS, Kneller RW, et al. Rising incidence of adenocarcinoma of the esophagus and gastric cardia. JAMA 1991;265:1287-9.
13. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
14. Devesa SS, Blot WJ, Fraumeni JF Jr. Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer 1998;83:2049-53.
15. Hansen S, Wiig JN, Giercksky KE, et al. Esophageal and gastric carcinoma in Norway 1958-1992: incidence time trend variability according to morphological subtypes and organ subsites. Int J Cancer 1997;71:340-4. [Crossref] [PubMed]
16. Nakauchi M, Walch HS, Nussenzweig S, et al. Genomic Landscape of Adenocarcinomas Across the Gastroesophageal Junction: Moving on From the Siewert Classification. Ann Surg 2025;281:989-96. [Crossref] [PubMed]
17. Siewert JR, Feith M, Stein HJ. Biologic and clinical variations of adenocarcinoma at the esophago-gastric junction: relevance of a topographic-anatomic subclassification. J Surg Oncol 2005;90:139-46; discussion 146. [Crossref] [PubMed]
18. Patel MA, Kratz JD, Lubner SJ, et al. Esophagogastric Cancers: Integrating Immunotherapy Therapy Into Current Practice. J Clin Oncol 2022;40:2751-62. [Crossref] [PubMed]
19. Grieb BC, Agarwal R. HER2-Directed Therapy in Advanced Gastric and Gastroesophageal Adenocarcinoma: Triumphs and Troubles. Curr Treat Options Oncol 2021;22:88. [Crossref] [PubMed]
20. Walch HS, Borpatragohain R, Jee J, et al. Clinical Implications of The Cancer Genome Atlas Molecular Classification System in Esophagogastric Cancer. Clin Cancer Res 2025;31:1912-21. [Crossref] [PubMed]
21. Janjigian YY, Al-Batran SE, Wainberg ZA, et al. Perioperative Durvalumab in Gastric and Gastroesophageal Junction Cancer. N Engl J Med 2025;393:217-30. [Crossref] [PubMed]
22. Shitara K, Van Cutsem E, Gümüş M, et al. Trastuzumab Deruxtecan or Ramucirumab plus Paclitaxel in Gastric Cancer. N Engl J Med 2025;393:336-48. [Crossref] [PubMed]
23. Yu J, Mehta R. Biomarker-Driven Approach to the Treatment of Metastatic Gastric or Gastroesophageal Adenocarcinoma. J Natl Compr Canc Netw 2025;23:e257036. [Crossref] [PubMed]
24. Siewert JR, Stein HJ. Classification of adenocarcinoma of the oesophagogastric junction. Br J Surg 1998;85:1457-9. [Crossref] [PubMed]
25. Bang YJ, Van Cutsem E, Feyereislova A, et al. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet 2010;376:687-97. [Crossref] [PubMed]
26. Peng Z, Zhang X, Liang H, et al. Atezolizumab and Trastuzumab Plus Chemotherapy for ERBB2-Positive Locally Advanced Resectable Gastric Cancer: A Randomized Clinical Trial. JAMA Oncol 2025;11:619-24. [Crossref] [PubMed]
27. Al-Batran SE, Homann N, Pauligk C, et al. Perioperative chemotherapy with fluorouracil plus leucovorin, oxaliplatin, and docetaxel versus fluorouracil or capecitabine plus cisplatin and epirubicin for locally advanced, resectable gastric or gastro-oesophageal junction adenocarcinoma (FLOT4): a randomised, phase 2/3 trial. Lancet 2019;393:1948-57.
28. Lin C, Wu J, Ding S, et al. Impact of Prior Cancer History on the Clinical Outcomes in Advanced Breast Cancer: A Propensity Score-Adjusted, Population-Based Study. Cancer Res Treat 2020;52:552-62. [Crossref] [PubMed]
29. Liu J, Chen Y, Zhan X, et al. Effect of prior cancer history on survival of patients with esophageal carcinoma: a propensity score matching, population-based study. J Thorac Dis 2022;14:979-94. [Crossref] [PubMed]
30. Laccetti AL, Pruitt SL, Xuan L, et al. Effect of prior cancer on outcomes in advanced lung cancer: implications for clinical trial eligibility and accrual. J Natl Cancer Inst 2015;107:djv002. [Crossref] [PubMed]
31. Chen Y, Sun R, Liu W. Impact of a previous cancer history on the overall survival of patients with primary gastric cancer: A SEER population-based study. Eur J Surg Oncol 2022;48:2159-65. [Crossref] [PubMed]
32. Wen L, Yu K, Lu H, et al. Impact of prior cancer history on survival of patients with gastric cancer. Eur J Surg Oncol 2021;47:2286-94. [Crossref] [PubMed]
33. Zhang H, Wang W, Cheng Y, et al. Adenocarcinomas of the esophagogastric junction: experiences at a single institution in China. World J Surg Oncol 2013;11:155. [Crossref] [PubMed]
34. Zhou Z, Zhang H, Xu Z, et al. Nomogram predicted survival of patients with adenocarcinoma of esophagogastric junction. World J Surg Oncol 2015;13:197. [Crossref] [PubMed]
35. Lai H, Zheng J, Zhou G, et al. Prognostic value of a new staging system for adenocarcinoma of the esophagogastric junction after preoperative radiation therapy. Asian J Surg 2023;46:2919-22. [Crossref] [PubMed]
36. Lai H, Zheng J, Zhou G, et al. Clinical characteristics and prognostic outcomes for adenocarcinoma of esophagogastric junction in early-onset patients: a population-based appraisal. J Cancer Res Clin Oncol 2023;149:14941-52. [Crossref] [PubMed]
37. Lai H, Zheng J, Li Y. Lymph node evaluation after neoadjuvant therapy for patients with adenocarcinoma of the esophagogastric junction. Asian J Surg 2023;46:2923-5. [Crossref] [PubMed]
38. Zhou H, Huang Y, Qiu Z, et al. Impact of prior cancer history on the overall survival of patients newly diagnosed with cancer: A pan-cancer analysis of the SEER database. Int J Cancer 2018;143:1569-77. [Crossref] [PubMed]