Comparison of Molecular-Genetic Concordance of the Primary Tumor and Brain Metastases of Gastroesophageal Cancers

Malignancies of the esophagus, gastroesophageal junction, and stomach, collectively referred to as gastroesophageal cancers, account for a substantial proportion of cancer incidence and mortality globally. The development of brain metastases (BM) in these patients, once considered an exceedingly rare event with incidences estimated at less than 1-2% in early case series, is now recognized with increasing frequency. This increasing frequency is largely attributed to advances in systemic therapy, which have led to improved control of extracranial disease and prolonged patient survival, as well as to improved neuroimaging, which has increased the detection of previously asymptomatic lesions, thereby unmasking the brain as a common sanctuary site for metastatic spread.

Despite this increasing recognition, the molecular-genetic landscape of BM from gastroesophageal cancers remains critically understudied, with limited data on key predictive biomarkers such as HER2, MSI, PD-L1, and CLDN18.2 in paired primary and metastatic samples. Nonetheless, the prognosis for patients with gastroesophageal cancer brain metastases has not improved over recent decades, with median survival still measured in months.

To address this critical knowledge gap, the international GASTROBRAIN study (ClinicalTrials.gov ID: NCT07448493) was previously initiated, which established a large multi-institutional retrospective cohort of over 230 patients with brain metastases from gastric and esophageal cancer, with comprehensive clinicopathological and treatment data. As the next step, archival histological material was systematically identified, collected, and centralized from patients with available paired formalin-fixed paraffin-embedded (FFPE) tissue samples of the primary tumor and corresponding BM for the translational GENCONCOR-2 study. This nested design will enable a robust investigation into the concordance of HER2, MSI, PD-L1 (CPS), and CLDN18.2 status in matched tumor pairs - an analysis that has not been previously reported.

Biomarker Assessment

1. HER2 status will be assessed by immunohistochemistry (IHC) using the SP3 antibody clone (DAKO) on the Ventana GX platform with the Ventana OptiView detection system. Results will be evaluated according to the ASCO-CAP guidelines. For gastric, gastroesophageal, and esophageal adenocarcinoma samples, HER2 IHC scoring will follow the modified criteria established for upper gastrointestinal cancers. For surgical specimens, 3+ positivity is defined as strong complete, basolateral or lateral membranous staining in ≥ 10% of tumor cells; for biopsy specimens, strong membranous staining in a cluster of at least 5 tumor cells is considered positive irrespective of the percentage of stained tumor cells. Cases with IHC 2+ (weak-to-moderate complete, basolateral or lateral membranous staining in ≥ 10% of tumor cells for surgical specimens, or in a tumor cell cluster for biopsies) will be considered equivocal and will undergo HER2 gene copy number assessment by in situ hybridization (ISH), including FISH, CISH, or SISH. HER2 positivity by ISH will be defined as a HER2/CEP17 ratio ≥ 2.0; or, if the ratio is < 2.0, an average HER2 gene copy number ≥ 6.0 signals per cell. Cases with IHC 2+ and a HER2/CEP17 ratio < 2.0 with an average HER2 gene copy number between ≥ 4.0 and < 6.0 signals per cell will be considered indeterminate. In such indeterminate cases, a second reviewer will be consulted, and retesting on additional tumor material may be considered.
2. PD-L1 status will be assessed by IHC using the DAKO 22C3 antibody clone on the Dako Link48 platform with the Dako EnVision Flex detection system. External positive control will be tonsil tissue. Results will be evaluated according to the test system manufacturer's recommendations. PD-L1 expression will be reported as the Combined Positive Score (CPS), defined as the number of PD-L1-stained cells (tumor cells, lymphocytes, macrophages) divided by the total number of viable tumor cells, multiplied by 100.
3. MSI status will be determined by IHC or PCR. For IHC assessment, loss of expression of mismatch repair proteins (MLH1, MSH2, MSH6, PMS2) will be evaluated. For PCR-based analysis, microsatellite instability will be assessed using five mononucleotide repeat markers (BAT25, BAT26, NR21, NR24, NR27).
4. CLDN18.2 expression will be assessed by IHC using the VENTANA CLDN18 (43-14A) assay on the Ventana platform. Positive expression will be defined as moderate-to-strong (2+/3+) complete, basolateral, or lateral membranous staining in ≥ 75% of viable tumor cells, in accordance with established criteria from clinical trials and current clinical guidelines. This threshold will be applied uniformly to all samples, including both primary tumors and brain metastases, from patients with gastric, gastroesophageal junction, and esophageal adenocarcinomas.
5. If funding becomes available, exploratory next-generation sequencing (NGS) will be performed on paired tumor samples to identify additional genomic alterations and investigate their concordance between primary tumors and brain metastases.

The primary objective of this study is to evaluate, in a large real-world cohort, the overall molecular discordance rate (%) between primary gastroesophageal cancers and their matched brain metastases - defined as the proportion of cases with discordant biomarker status relative to the total number of analyzed paired samples. In addition to the overall discordance rate, the discordance rate (%) will be analyzed separately for each biomarker (HER2, MSI, PD-L1 (CPS), and CLDN18.2). Secondary endpoints include:

1. Overall Survival (OS): Defined as the time from the date of brain metastasis (BM) diagnosis to the date of death from any cause or last follow-up (censored).
2. Time to Intracranial Progression (TTIP): Defined as the time from the date of initial gastric and esophageal cancer diagnosis to the date of first BM detection. Based on this interval, patients will be categorized into two groups:

• Synchronous BM: BM diagnosed either prior to or within 2 months (≤ 60 days) of the primary tumor diagnosis.
• Metachronous BM: BM diagnosed more than 2 months (> 60 days) after the primary tumor diagnosis.

1. Central Nervous System Progression-Free Survival (CNS-PFS): Defined as the time from the date of first local treatment for BM to the date of subsequent intracranial progression or last instrumental follow-up (censored). Subsequent intracranial progression includes:

• Continued growth of the treated lesion (≤ 6 months thereof);
• Local recurrence of the treated lesion (> 6 months thereof);
• Development of new distant intracranial lesions. Exploratory objective: Subject to availability of funding, next-generation sequencing (NGS) will be performed on paired samples to explore the concordance of a broader panel of genomic alterations between primary tumors and their matched brain metastases.

Statistical Analysis. All statistical analyses will be performed using IBM SPSS Statistics (version 29.0) and STATA (version 17.0, StataCorp LLC). A two-sided p-value < 0.05 will be considered statistically significant.

• Descriptive Statistics. Categorical variables will be presented as absolute frequencies (n) and relative frequencies (%). Continuous variables will be assessed for normality. Normally distributed variables will be presented as mean with standard deviation (SD); non-normally distributed variables will be presented as median with interquartile range (IQR). The frequency of missing data will be reported for all variables.
• Discordance and Concordance Analysis. The primary endpoint is the overall molecular discordance rate (%) - defined as the proportion of cases with discordant biomarker status between the primary tumor and its matched brain metastasis relative to the total number of analyzed paired samples. In addition to the overall discordance rate, the discordance rate (%) will be analyzed separately for each biomarker (HER2, MSI, PD-L1 (CPS), and CLDN18.2). Furthermore, concordance will be evaluated using overall percent agreement and Cohen's kappa coefficient (κ) with 95% confidence intervals. Kappa values will be interpreted as follows: ≤ 0 = poor agreement, 0.01-0.20 = slight agreement, 0.21-0.40 = fair agreement, 0.41-0.60 = moderate agreement, 0.61-0.80 = substantial agreement, and 0.81-1.00 = almost perfect agreement.
• Survival Analysis. Survival curves will be estimated using the Kaplan-Meier method. Median survival times with 95% confidence intervals (CI) will be reported. Comparison of survival curves between groups (e.g., concordant vs. discordant cases, biomarker-positive vs. biomarker-negative) will be performed using the log-rank test.
• Univariable and Multivariable Analysis. Univariable Cox proportional hazards regression will be performed to identify potential prognostic factors associated with survival outcomes (OS, CNS-PFS). Variables with p < 0.10 on univariable analysis, as well as clinically relevant factors regardless of significance, will be entered into multivariable Cox proportional hazards regression models to identify independent prognostic factors. The proportional hazards assumption will be tested. Results will be presented as hazard ratios (HR) with 95% CI.
• Association with Clinicopathological Data. Biomarker status and discordance patterns will be correlated with clinicopathological variables and treatment modalities using chi-square test or Fisher's exact test for categorical variables, and t-test or Mann-Whitney U test for continuous variables, as appropriate. These analyses will leverage the comprehensive clinicopathological dataset established within the parent GASTROBRAIN cohort.
• Handling of Missing Data. The proportion of missing data will be reported for all variables. Patterns of missingness will be explored. Given the ancillary nature of this translational study, only cases with complete paired biomarker data will be included in this analysis.