Future Directions in Integrating Genomics and Immunotherapy for Gastrointestinal Cancers

A New Era for Precision Immunotherapy in GI Cancers

Gastrointestinal (GI) cancers represent a formidable global health challenge, accounting for 26% of all cancer incidence and 35% of cancer-related deaths worldwide. Colorectal, pancreatic, and liver cancers lead as the top causes of GI cancer mortality in the United States. This immense burden underscores an urgent need for novel, integrated approaches that combine insights from genomics with the power of immunotherapy.

The Path from Broad Treatment to Biomarker-Guided Therapy

The transition from conventional therapies to precision immuno-oncology is driven by the recognition that GI cancers are not a single entity. They encompass a spectrum of diseases with distinct molecular profiles, tumor microenvironments, and immune landscapes. While immune checkpoint inhibitors (ICIs) have revolutionized oncology, their efficacy in GI cancers has been largely confined to specific molecular subgroups, highlighting both the promise and the limitations of current immunotherapy.

A landmark success lies in the treatment of mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-H) tumors. In colorectal cancer, this genomic feature renders tumors highly immunogenic, leading to deep and durable responses with PD-1 blockade. For patients with dMMR/MSI-H metastatic colorectal cancer, anti-PD-1 monotherapy or dual PD-1/CTLA-4 blockade has become a standard first-line therapy, often replacing chemotherapy. In the neoadjuvant setting, PD-1 blockade achieves clinical complete response rates of 75–100% in locally advanced dMMR rectal cancer, enabling watch-and-wait strategies that avoid surgery and its lifelong consequences.

These successes, however, illuminate a stark contrast: only about 10–20% of all GI cancer patients respond to current ICIs. The vast majority of pancreatic ductal adenocarcinomas are immune-quiescent, and microsatellite-stable (MSS) colorectal cancers are notoriously resistant. This clinical reality is driving a shift toward biomarker-driven patient selection and rationally designed combination therapies.

Beyond Established Biomarkers: A Genomic Roadmap

The integration of genomics into immunotherapy is expanding beyond MSI and PD-L1 testing. Tumor mutational burden (TMB) and the composition of the tumor microenvironment (TME) are increasingly used to stratify patients. For instance, the COMPASSION-15 trial showed that cadonilimab (an anti-PD-1/CTLA-4 bispecific antibody) provided survival benefit in gastroesophageal adenocarcinoma even in patients with low PD-L1 expression, suggesting that dual checkpoint blockade can overcome the limitations of PD-L1-based selection.

Emerging evidence suggests that specific genomic alterations can predict immunotherapy outcomes. The development of the Gastrointestinal Immune Prognostic Signature (GIPS), a six-gene model using mutations in RNF43, CREBBP, CDKN2A, TP53, SPEN, and NOTCH3, exemplifies this trend. In validation cohorts, GIPS-high patients had significantly longer overall survival and higher rates of durable clinical benefit compared to GIPS-low patients. Importantly, GIPS outperformed or complemented existing biomarkers like PD-L1 and TMB. When combined, GIPS-high and TMB-high identified a subgroup with the best outcomes, pointing to a future of combinatorial genomic biomarkers.

The Tumor Microenvironment as a Therapeutic Target

The complex interplay between tumor cell-intrinsic genomics and the TME is central to immunotherapy resistance. The TME includes cancer-associated fibroblasts (CAFs), tumor-associated macrophages (TAMs), and the gut microbiome. CAF plasticity is subtype-specific: in pancreatic cancer, depleting myofibroblastic CAFs paradoxically accelerated tumor progression, teaching a cautionary tale about blunt targeting.

However, other TME components offer actionable targets. M2-polarized TAMs, which dominate the immune landscape in many GI cancers, promote immunosuppression and resistance. Strategies to reprogram them—for example, by blocking CSF1R—can convert a suppressive TME into an immune-responsive one. The gut microbiome also plays a critical role: Fusobacterium nucleatum in colorectal cancer promotes chemoresistance, while specific intratumoral bacteria like Pseudoxanthomonas in pancreatic cancer correlate with long-term survival and CD8+ T cell infiltration. Modulating the microbiome, through antibiotics or fecal microbiota transplantation, represents a frontier for combination with immunotherapy.

Forging the Path Forward

The future of GI cancer immunotherapy lies in the convergence of genomics, TME profiling, and multi-omics analysis. A "three-strata" framework is emerging: highly immunogenic tumors (dMMR/MSI-H) can pursue chemotherapy-free, immunotherapy-centric strategies; intermediate tumors (e.g., gastroesophageal cancers) benefit from chemo-immunotherapy guided by PD-L1 scores; poorly immunogenic tumors (e.g., MSS colorectal, pancreatic) require rationally designed combinations that target multiple immune defects simultaneously.

The CAPability-01 trial demonstrated that a triplet of sintilimab, a histone deacetylase inhibitor, and an anti-VEGF agent improved response rates in refractory MSS colorectal cancer from 13% to 44%. This underscores that overcoming resistance requires attacking the problem from multiple angles—enhancing T cell infiltration and blocking immunosuppressive pathways simultaneously.

Ultimately, the integration of genomics and immunotherapy for GI cancers is no longer a distant aspiration but an active, evolving clinical reality. The challenge now is to validate emerging biomarkers, standardize multi-omics platforms, and design adaptive clinical trials that match the complexity of these diseases. With continued progress, precision immunotherapy has the potential to transform outcomes for the millions of patients affected by gastrointestinal cancers worldwide. In the following sections, we explore these combination strategies, novel immunotherapeutic modalities, and the predictive tools that will define this new era.

Current Progress and Persistent Challenges in GI Immunotherapy

Yes, immunotherapy is now an established treatment for gastric cancer. Multiple PD-1/PD-L1 checkpoint inhibitors, including nivolumab and pembrolizumab, have shown significant efficacy, particularly in patients with PD-L1 CPS ≥5 or microsatellite instability-high (MSI-H) tumors. Additionally, newer agents like dostarlimab and targeted antibodies (trastuzumab deruxtecan) have expanded the therapeutic arsenal. Treatment selection is increasingly guided by molecular profiling, which helps identify patients most likely to benefit from these approaches.

Overview of challenges: most GI cancers remain resistant to single-agent checkpoint inhibitors

Despite the success in specific molecular subtypes, the broader landscape of gastrointestinal cancers presents considerable obstacles.

• Microsatellite stable (MSS) colorectal cancer: This subtype, representing the vast majority of CRC cases, is notoriously resistant to single-agent PD-1/PD-L1 blockade, with objective response rates of only 0–2%. Multiple phase 3 trials combining checkpoint inhibitors with other agents have failed to demonstrate a survival benefit.
• Pancreatic ductal adenocarcinoma (PDAC): Most PDACs are immunologically "cold" and resistant to single-agent checkpoint blockade due to their low tumor mutational burden and highly immunosuppressive tumor microenvironment characterized by abundant cancer-associated fibroblasts, M2-polarized macrophages, and dense stroma.
• Liver and biliary cancers: These tumors also exhibit variable immunogenicity and often fail to respond to monotherapy.

The underlying causes of resistance include low neoantigen burden, lack of T-cell infiltration, and the presence of multiple immunosuppressive mechanisms within the tumor microenvironment. These challenges highlight the need for rationally designed combination strategies that target not only immune checkpoints but also other components of the tumor microenvironment, such as cancer-associated fibroblasts, macrophages, and metabolic pathways. Ongoing research focuses on integrating genomic profiling, tumor microenvironment characterization, and novel immunotherapeutic modalities (e.g., CAR-T cells, bispecific antibodies, cancer vaccines) to convert resistant tumors into responsive ones. The development of robust predictive biomarkers beyond PD-L1 and MSI status remains a critical priority for advancing immunotherapy across all GI cancer types.

Success Stories: Where Genomics Already Guides Treatment

How successful is immunotherapy for bowel cancer?

For a specific subset of bowel cancer patients defined by their genomic profile, the results have been transformative. A recent phase II trial investigated the immunotherapy drug pembrolizumab as a pre-surgical (neoadjuvant) treatment. The outcomes were striking: all signs of cancer were eliminated in 59% of patients with mismatch repair deficient (dMMR) / microsatellite instability-high (MSI-H) bowel cancer.

The success of neoadjuvant immunotherapy in dMMR/MSI‑H colorectal cancer

This success extends beyond complete responses. For the remaining patients who did not achieve a complete pathological response (pCR), any residual cancer was successfully removed during surgery. Remarkably, all 32 participants in the trial were rendered cancer-free after the combined treatment, with a median cancer-free period of 9.7 months. This represents a dramatic improvement over standard chemotherapy, which typically achieves a pCR in less than 5% of these patients.

Patient Group	Pathological Complete Response (pCR)	Outcome
dMMR/MSI‑H (Pembrolizumab)	59%	All patients cancer-free after surgery; avoided post-operative chemo.
Standard Chemotherapy (Historical)	< 5%	Majority require follow-up treatments.

In essence, for patients whose tumors carry these specific genomic signatures, immunotherapy has shifted the paradigm from a systemic, often toxic treatment to a highly effective, targeted intervention that can eliminate the need for post-surgical chemotherapy entirely.

Beyond Traditional Biomarkers: The Rise of Genomic Signatures

Limitations of Current Biomarkers

Current biomarkers like MSI-H, PD-L1, and TMB are essential but have notable shortcomings. MSI-H occurs in only 0–5% of metastatic gastrointestinal cancers, limiting its reach. PD-L1 expression yields inconsistent results across trials and suffers from interobserver variability. TMB lacks a consensus cutoff and treats all mutations equally, ignoring that certain mutations (e.g., TP53, STK11) have different immunologic impacts.

The Gastrointestinal Immune Prognostic Signature

The novel Gastrointestinal Immune Prognostic Signature (GIPS) addresses these gaps. This six-gene panel integrates mutations in RNF43, CREBBP, CDKN2A, TP53, SPEN, and NOTCH3 into a cost-effective assay. GIPS-high patients had significantly better survival (median OS 31 vs. 10 months; HR 0.40). The signature outperformed PD-L1 and MSI-H in predicting outcomes and could be combined with TMB for even finer patient stratification.

Biomarker	Main Limitation	GIPS Advantage
MSI-H	Very low prevalence (0–5%)	Applicable to broader patient population
PD-L1	High interobserver variability, controversial results	Standardized gene panel, consistent measure
TMB	No consensus cutoffs, all mutations weighted equally	Captures distinct immunologic impact of specific genes
GIPS	Requires further prospective validation	Cost-effective, outperforms existing markers alone

Systematic Approaches: Stratifying Tumors by Immunogenicity

• Stratum 1 (Highly Immunogenic): This includes tumors like dMMR/MSI‑H colorectal cancer. These tumors have a high tumor mutational burden and pre‑existing T‑cell infiltrates. The focus is on chemotherapy‑free, immunotherapy‑centric strategies, often using immune checkpoint inhibitors alone or in combination.
• Stratum 2 (Intermediate Immunogenicity): This group includes gastroesophageal cancers. Here, the established standard of care is chemo‑immunotherapy combinations, which are typically effective for patients with a PD‑L1 CPS ≥5.
• Stratum 3 (Poorly Immunogenic or "Cold"): This includes tumors such as pMMR/MSS colorectal cancer and the majority of pancreatic ductal adenocarcinomas (PDAC). These tumors are highly resistant to single‑agent immunotherapy and require rationally designed combination therapies.

What Novel Targets and Strategies Are Emerging for "Cold" Tumors?

For immunologically "cold" tumors in Stratum 3, research focuses on converting the immunosuppressive tumor microenvironment (TME) into a state that supports an immune response. Key targets and approaches include:

Target / Strategy	Mechanism of Action	Example in Clinical Development
CSF1R Inhibition	Reprograms tumor‑associated macrophages (TAMs) from an immunosuppressive (M2) to an antitumor (M1) state.	Nivolumab (anti‑PD‑1) + cabiralizumab (anti‑CSF1R) for PDAC.
TGFβ Inhibition	Blocks a key cytokine that promotes an immune‑excluded TME and inhibits T‑cell activity.	Checkpoint blockade combined with a TGFβ inhibitor in preclinical PDAC models.
HDAC + VEGF Inhibition	Synergistically reactivates T‑cell trafficking and reduces immunosuppressive signaling within the tumor.	Sintilimab (anti‑PD‑1) + chidamide (HDACi) + bevacizumab (anti‑VEGF) in refractory pMMR/MSS mCRC (CAPability‑01 trial).
Dual Checkpoint Blockade	Simultaneously targets multiple inhibitory pathways (e.g., PD‑1 and CTLA‑4) to revive T‑cell function.	Nivolumab + ipilimumab is being explored for PDAC and esophageal squamous cell carcinoma.
TAM Kinase Inhibition	Blocks the activity of TYRO3, AXL, and MERTK (TAM) kinases on tumor cells and macrophages, counteracting liver‑specific immunosuppression.	Zanzalintinib + atezolizumab in pMMR/MSS mCRC with liver metastases (STELLAR‑303 trial).

This strategic, biology‑driven approach is essential for making progress against the most challenging gastrointestinal malignancies.

Novel Frontiers: CAR-T, Vaccines, and Multi-Omics Integration

Novel Frontiers: CAR-T, Vaccines, and Multi-Omics Integration

Novel cell therapies are entering GI oncology. CAR-T cells targeting Claudin18.2 (CT041) achieved an objective response rate of 48.6% in later-line disease, while EphA2-targeted CAR-T shows promise against esophageal squamous cell carcinoma. Personalized neoantigen vaccines, such as the mRNA-based autogene cevumeran, induce CD8+ T cells with an estimated lifespan of 7.7 years and prolonged recurrence-free survival in pancreatic cancer.

Artificial intelligence and multi-omics are transforming patient selection. Deep learning models classify tumor microenvironment subtypes (AUC 0.94) and predict MSI status from histology. Integration of genomics, transcriptomics, and metabolomics, combined with dynamic ctDNA monitoring, enables real-time treatment adaptation. The gastrointestinal immune prognostic signature (GIPS) exemplifies a genomic tool that, when combined with tumor mutational burden, improves patient stratification.

The path forward requires prospective validation of these integrated approaches. Combining genomic profiling, AI-driven analysis, and non-invasive monitoring will allow clinicians to rationally select between checkpoint inhibitors, CAR-T cells, vaccines, or oncolytic viruses, ultimately achieving personalized immunotherapy for each GI cancer patient.

A Personalized Future Rooted in Hope and Science

The field is moving beyond a one-size-fits-all approach to immunotherapy. The future lies in genomics-driven, personalized strategies that tailor treatment to the unique molecular and immune profile of each patient's tumor. This shift is already underway, as signatures like GIPS and the three-strata framework begin to guide clinical decisions, moving from simply treating the cancer to precisely targeting its vulnerabilities.

Continued research and well-designed clinical trials remain essential to overcome resistance and expand the benefits of immunotherapy to more patients. Exploring mechanisms of immune evasion in pancreatic cancer, validating composite biomarkers like GIPS, and testing rational combinations in microsatellite-stable colorectal cancer are critical next steps. The goal is to transform more gastrointestinal cancers from immunologically "cold" to "hot," turning scientific insight into durable patient outcomes.