ASCO Conference (GI): 2026, Chicago,IL.

HCC treatment is rapidly moving beyond the old model of surgery, ablation, TACE, and systemic therapy as separate compartments. Immunotherapy is now being tested earlier—before surgery, after surgery, and in combination with locoregional treatment. The rationale is strong: HCC often recurs after resection or ablation, and micrometastatic disease may already be present even when imaging suggests localized disease. Neoadjuvant immunotherapy may shrink tumors, increase immune priming, and help identify aggressive biology before surgery. Adjuvant immunotherapy aims to reduce recurrence after curative-intent resection or ablation. In intermediate-stage HCC, systemic therapy is increasingly being combined with TACE rather than waiting until TACE failure. EMERALD-3 is especially important because durvalumab/tremelimumab plus lenvatinib and TACE reduced the risk of progression or death compared with TACE alone in unresectable embolization-eligible HCC. The challenge is liver function. In HCC, treatment success is not only tumor response but also preservation of hepatic reserve. Patient selection should consider BCLC stage, Child-Pugh score, portal hypertension, transplant eligibility, viral hepatitis control, and bleeding risk. Practical pearl: HCC immunotherapy is moving earlier, but every decision must balance oncologic benefit against liver-function safety.

HER2 is no longer only a breast cancer biomarker. It is clinically important in gastric/GEJ cancer, colorectal cancer, biliary tract cancer, and rarely pancreatic cancer. In gastric and GEJ adenocarcinoma, trastuzumab with platinum-fluoropyrimidine chemotherapy remains a historical first-line standard for HER2-positive disease. The field has moved beyond trastuzumab alone. Current strategies include trastuzumab plus immunotherapy, antibody-drug conjugates such as trastuzumab deruxtecan, bispecific antibodies, and dual HER2 blockade. Trastuzumab deruxtecan has become a major later-line option in HER2-positive gastric cancer and is being explored across GI tumors. In colorectal cancer, HER2 amplification is usually seen in RAS wild-type tumors and may act as a resistance mechanism to anti-EGFR therapy. HER2-targeted combinations have shown activity in metastatic colorectal cancer, especially in HER2-positive, RAS wild-type disease. HER2 expression is also clinically relevant in biliary tract cancers, where HER2-directed therapies are increasingly entering precision oncology pathways. HER2 testing should be done early in metastatic gastric/GEJ cancer and considered in molecular profiling panels for metastatic colorectal and biliary tract cancers. For GastroAGI readers, HER2 should be viewed as a cross-GI actionable biomarker, not a single-disease marker. Practical pearl: HER2-positive GI cancers need disease-specific interpretation because HER2 biology, testing thresholds, and treatment benefit differ between gastric, colorectal, and biliary cancers.

CLDN18.2 has become one of the most important new targets in gastric and gastroesophageal junction adenocarcinoma. It is a tight-junction protein normally hidden in gastric mucosa but exposed and overexpressed in many gastric cancers, making it a useful therapeutic target. Zolbetuximab, a CLDN18.2-directed antibody, was approved by the FDA in October 2024 with fluoropyrimidine- and platinum-based chemotherapy for first-line treatment of HER2-negative, CLDN18.2-positive, unresectable or metastatic gastric/GEJ adenocarcinoma. The approval was based on the SPOTLIGHT and GLOW phase 3 trials, where adding zolbetuximab improved progression-free and overall survival compared with chemotherapy alone. Importantly, FDA also approved the VENTANA CLDN18 companion diagnostic assay to identify eligible patients. Common toxicities include nausea, vomiting, diarrhea, neutropenia, and infusion-related symptoms; therefore, supportive care and patient selection matter. For GastroAGI readers, CLDN18.2 should now be part of the initial biomarker panel in advanced gastric/GEJ cancer, along with HER2, PD-L1 CPS, MSI/MMR, and NGS where available. Practical pearl: CLDN18.2 has transformed gastric cancer from chemotherapy-plus-immunotherapy alone into a true biomarker-layered disease.

MRD means microscopic residual cancer that remains after apparently curative treatment and is not visible on imaging. In GI oncology, ctDNA is becoming the most practical MRD tool. The strongest evidence is in colorectal cancer, where postoperative ctDNA positivity predicts a very high recurrence risk. Its main applications are: selecting patients for adjuvant chemotherapy, de-escalating therapy in ctDNA-negative low-risk patients, detecting recurrence earlier than CT scan, and monitoring resistance in metastatic disease. In stage II colon cancer, ctDNA-guided strategies have shown that some patients may safely avoid chemotherapy without compromising outcomes, but broad survival-changing evidence is still evolving. In stage III colon cancer and rectal cancer, trials such as CIRCULATE, COBRA, DYNAMIC, and BESPOKE-type studies are helping define how much treatment should be intensified or reduced. In gastric, pancreatic, and biliary cancers, ctDNA is promising but less mature than in colorectal cancer. The key GastroAGI message: ctDNA is highly prognostic, but ctDNA-guided treatment is not yet universally standard in all GI cancers. Clinicians should avoid using ctDNA alone. It must be integrated with TNM stage, pathology, imaging, surgical margins, MSI status, patient fitness, and recurrence risk. Practical pearl: ctDNA is becoming the molecular version of surveillance imaging—but the next step is proving that acting on it improves survival.

Oncology trials have traditionally focused on overall survival, progression-free survival, response rate, and toxicity. ASCO 2026 includes a Clinical Science Symposium on choosing the right patient-centered endpoints in modern clinical trials , showing a major shift toward outcomes that matter directly to patients. Patient-centered endpoints include quality of life, symptom control, functional independence, treatment-free interval, steroid-free survival, time at home, financial toxicity, caregiver burden, and patient-reported outcomes. This is especially important in GI oncology because many treatments prolong survival but may also cause diarrhea, fatigue, neuropathy, anorexia, pain, sarcopenia, hepatic decompensation, or repeated hospital visits. In pancreatic cancer, a small survival gain may be meaningful only if pain, nutrition, and performance status are preserved. In HCC, endpoints must account for liver function, not just tumor response. In colorectal cancer, nonoperative management, ctDNA-guided de-escalation, and organ preservation require endpoints beyond radiological response. Regulators, clinicians, and patients increasingly recognize that “living longer” and “living better” must be measured together. For GastroAGI readers, this topic is conceptually very important because future GI oncology trials will not be judged only by hazard ratios. The clinical pearl: a modern cancer trial should answer two questions: does the treatment extend life, and does it protect the life the patient wants to live?

AI in oncology has moved from curiosity to practical clinical infrastructure. ASCO 2026 includes multiple AI-focused sessions, including clinical application of evolving technologies and AI , AI for smarter and more accessible cancer care , and integrating AI into oncology practice . In GI oncology, AI can support screening, endoscopy quality, radiology interpretation, pathology quantification, molecular tumor boards, clinical trial matching, toxicity prediction, and patient education. AI is especially promising in colorectal cancer screening, where colonoscopy AI improves adenoma detection, and in digital pathology, where models can identify MSI-like patterns, tumor budding, lymphovascular invasion, and immune microenvironment features. In research, AI can accelerate trial screening and reduce workload. Recent oncology AI systems have shown high accuracy for clinical trial eligibility prescreening from clinical documents. For low- and middle-income countries, AI may help address shortage of oncology specialists, but only if tools are validated locally. The risks are real: hallucination, bias, poor explainability, medicolegal uncertainty, data privacy issues, and overdependence on unvalidated tools. For GastroAGI readers, AI should be seen as an “oncology co-pilot,” not an autonomous decision maker. The practical pearl: AI is most useful when it improves workflow, accuracy, access, or consistency—but final treatment decisions must remain clinician-led and evidence-based.

Early-onset cancer is becoming a major public health and GI oncology issue, particularly early-onset colorectal cancer. ASCO 2026 includes the ASCO/ASPO joint session “The Coming Wave of Early-Onset Cancer,” highlighting rising incidence, risk factors, and prevention strategies. Early-onset GI cancers include colorectal, pancreatic, gastric, esophageal, biliary, appendiceal, and neuroendocrine cancers. Recent reports suggest early-onset GI cancers rose substantially between 2010 and 2019, with disproportionate effects in some racial, ethnic, and sex groups. Early-onset colorectal cancer is especially concerning because many patients present with advanced-stage disease, often after delayed symptom recognition. Risk factors are likely multifactorial: obesity, sedentary lifestyle, ultra-processed foods, alcohol, microbiome changes, antibiotic exposure, sleep disruption, environmental exposures, and hereditary syndromes. However, no single cause explains the epidemic. For GastroAGI readers, this topic is highly relevant because gastroenterologists are central to symptom recognition, colonoscopy access, family-history assessment, and prevention messaging. Red-flag symptoms in younger adults—rectal bleeding, iron-deficiency anemia, unexplained weight loss, persistent bowel habit change, and abdominal pain—should not be dismissed as IBS or hemorrhoids without evaluation. The clinical pearl: early-onset GI cancer is not rare enough to ignore anymore; prevention must begin before symptoms and diagnosis must not be delayed after symptoms.

GIST treatment has historically been one of the best examples of targeted therapy in GI oncology, beginning with imatinib for KIT/PDGFRA-driven disease. However, after imatinib failure, second-line progress has been slow for more than two decades. ASCO 2026 highlighted the phase 3 PEAK study of bezuclastinib plus sunitinib versus sunitinib alone in advanced GIST . Bezuclastinib is a selective KIT inhibitor designed to complement sunitinib activity and overcome resistance biology. Reported PEAK data showed median progression-free survival of approximately 16.5 months with bezuclastinib plus sunitinib versus 9.2 months with sunitinib alone , with higher response rates in the combination arm. This is clinically meaningful because second-line GIST has relied heavily on sunitinib since 2006, and durable improvement in this space has been difficult. If confirmed through regulatory review and mature publication, this combination could reshape the post-imatinib treatment sequence. Molecular profiling remains essential: KIT exon 11, KIT exon 9, PDGFRA D842V, SDH-deficient GIST, NF1-associated GIST, and BRAF/NTRK-driven rare GISTs require different thinking. For GastroAGI readers, PEAK is not simply a “new drug” story; it is a resistance-overcoming strategy. The key pearl: GIST is becoming a mutation- and resistance-stage–directed disease, not merely a line-of-therapy disease.

Nonmetastatic exocrine pancreatic cancer requires one of the most complex multidisciplinary decisions in GI oncology. The ASCO 2026 program includes a dedicated case-based panel on optimal multidisciplinary care for nonmetastatic exocrine pancreatic cancer , reflecting the importance of coordinated treatment sequencing. The first critical step is accurate staging: high-quality pancreas-protocol CT or MRI, CA 19-9 assessment, performance status, nutritional status, biliary drainage if needed, and early surgical review. The major treatment categories are resectable, borderline resectable, locally advanced unresectable, and occult metastatic disease. Modern management increasingly favors neoadjuvant systemic therapy, especially for borderline resectable and locally advanced disease, because it treats micrometastatic disease early and helps select patients who will truly benefit from surgery. FOLFIRINOX-based therapy remains a major backbone in fit patients; gemcitabine/nab-paclitaxel is used when FOLFIRINOX is unsuitable. Radiation or chemoradiation remains selective rather than universal, usually considered for persistent vascular involvement, local control, or conversion strategy. Surgery should be performed in experienced high-volume centers because margin status, vascular reconstruction, and postoperative outcomes strongly influence survival. For GastroAGI readers, the message is clear: pancreatic cancer should never be treated by one specialist alone. The practical pearl: the best outcomes come from early tumor board discussion before the first cycle of therapy , not after chemotherapy has already begun.

Neuroendocrine tumors are no longer managed as one uniform disease. Current treatment depends on tumor site, grade, Ki-67, somatostatin receptor expression, functional status, disease burden, growth rate, and prior therapy. The ASCO 2026 session on new treatment paradigms in NETs is important because NET therapy is rapidly moving from passive observation and somatostatin analogues toward structured sequencing of targeted therapy, PRRT, chemotherapy, and liver-directed therapy. Somatostatin analogues remain foundational for symptom control and antiproliferative benefit in well-differentiated somatostatin receptor–positive NETs. PRRT with lutetium-177 DOTATATE is now a key therapeutic pillar in progressive somatostatin receptor–positive gastroenteropancreatic NETs; its role is expanding earlier in treatment algorithms. Cabozantinib became an important new option after FDA approval in March 2025 for previously treated unresectable, locally advanced, or metastatic well-differentiated pancreatic and extra-pancreatic NETs, based on the phase 3 CABINET trial. The practical question is no longer “what drug works?” but “what is the correct sequence?” For GastroAGI readers, NET management should be framed around multidisciplinary planning: medical oncology, nuclear medicine, hepatobiliary surgery, interventional radiology, pathology, and endocrinology. The key clinical pearl: grade, receptor expression, and pace of disease should drive therapy selection. Slow-growing SSTR-positive disease may favor SSA/PRRT sequencing, while rapidly progressive or high-volume disease may require targeted therapy or chemotherapy.