Comprehensive genomic profiling (CGP) is increasingly used in GI oncology to identify molecular alterations that may expand treatment opportunities beyond routine biomarker testing. However, translating genomic findings into treatment remains challenging, particularly when tissue quality, patient condition, and access to matched therapies or clinical trials limit their clinical application.
A large real-world study from the Veneto Institute of Oncology-IRCCS in Padua, Italy, evaluated the feasibility, molecular yield, and clinical impact of CGP in patients with gastrointestinal cancers.
The study, titled “Performance and Clinical Benefit of Comprehensive Genomic Profiling of GI Tumors in a Comprehensive Cancer Center,” was published on August 12, 2026, in JCO Precision Oncology.
Authors: Giulia Maddalena, Valentina Angerilli, Federico Nichetti, Gianmarco Ricagno, Jessica Gasparello, Elena Mattiuzzo, Maria Caterina De Grandis, Elena Carcea, Carlotta Ceccon, Anna Roma, Eleonora Perissinotto, Sara Sperotto, Giacomo Di Paolo, Marta Sbaraglia, Marco Maruzzo, Francesca Bergamo, Angelo Paolo Dei Tos, Matteo Fassan, and Sara Lonardi.
Evaluating CGP in Routine GI Oncology Practice
This retrospective observational study included consecutive patients with advanced GI cancers treated at the Veneto Institute of Oncology between December 2019 and January 2022. Clinical and molecular data were updated through February 2024. Patients underwent CGP using the FoundationOne CDx panel, primarily within clinical trials or following evaluation by a Molecular Tumor Board.
The investigators assessed whether CGP could successfully generate molecular information, identify actionable genomic alterations beyond routine diagnostic testing, and provide additional opportunities for matched treatment. CGP was proposed for 1,450 patients with GI tumors. Of these, 1,310 samples, or 90.3%, were adequate for sequencing, while 140 samples, or 9.7%, were considered inadequate.
Among the inadequate samples, the most commonly reported problems were low tumor nuclei content in 47.1%, low DNA quality in 26.4%, and low DNA quantity in 14.3%. Failure was proportionally more frequent with biopsy specimens than with surgical resection samples. The investigators emphasized that specimen selection and preanalytic factors can therefore have a substantial impact on successful molecular profiling.
Even among samples considered adequate for sequencing, 252 of 1,310, or 19.2%, generated informative reports that were classified as low quality.
CGP Identified Additional Actionable Alterations
Of the 1,310 patients with adequate CGP results, 1,265 had advanced, stage IV GI cancers and formed the population for the main clinical analysis. CGP identified predefined actionable alterations not routinely assessed — defined by the investigators as “gain alterations” — in 355 of 1,265 patients, or 28.1%.
This corresponded to a 24.5% gain in molecular information across the overall study population. Among patients with gain alterations, 51.0% had colorectal cancer, 24.5% had hepato-biliary-pancreatic cancers, 18.6% had upper GI cancers, and 5.9% had tumors categorized as rare GI malignancies.
The most frequently identified gain alteration was a PIK3CA mutation, reported in 147 cases, representing 32.8% of identified alterations. High tumor mutational burden in microsatellite-stable or mismatch repair–proficient tumors was the second most common finding, accounting for 65 alterations, or 14.5%. Some patients had more than one potentially actionable genomic finding: 56 patients had two gain alterations and six had three.
Finding an Actionable Alteration Did Not Always Lead to Treatment
The study also highlighted a major challenge of precision oncology: identifying a target does not necessarily mean that a patient will receive a matched therapy. Clinical treatment opportunities were evaluated in 263 patients with actionable alterations after excluding patients who were lost to follow-up or had insufficient performance status for further treatment.
Among these patients, 55 received a targeted therapy based on the identified alteration. The most frequently actioned molecular findings included MSS/pMMR tumors with high TMB, MSI-high status in non-colorectal cancers, and ERBB2 amplifications.
Matched treatments were delivered through several routes. Overall, 38.2% were administered within clinical trials, 36.4% through institutional off-label prescriptions, and 25.5% through nominal access programs. Across the entire population of 1,265 patients with advanced GI cancers, the investigators reported that CGP had provided an additional treatment opportunity to approximately 4.5% of patients at the time of analysis.
Access Remained a Major Barrier
Among 283 patients with actionable alterations who did not receive a targeted treatment, the most common reason was the absence of an available clinical trial or off-label or nominal-use program, reported in 37.5%. Another 31.4% remained on an ongoing treatment or were under follow-up and could potentially become candidates for matched treatment later. Poor performance status prevented targeted therapy in 26.5%, while 4.6% did not meet clinical trial inclusion or exclusion criteria.
These findings underline a central limitation of genomic medicine: molecular actionability and clinical actionability are not equivalent. The availability of clinical trials, expanded-access programs, and the patient’s clinical condition ultimately determine whether a genomic finding can be translated into treatment.
Outcomes With CGP-Guided Therapy
After a median follow-up of 22.2 months, when survival was assessed accounting for the time of CGP testing, median overall survival was 12.0 months among patients without gain alterations, 10.9 months among those with gain alterations who did not receive matched treatment, and 26.4 months among patients who received treatment directed at an actionable alteration. When survival was evaluated from the diagnosis of metastatic disease, there was no statistically significant difference between treated and untreated patients with gain alterations, with an HR of 0.84 (95% CI, 0.58–1.22; P = .352).
However, when outcomes were assessed accounting for the time of CGP testing, patients with gain alterations who received tailored treatment had longer overall survival than patients without gain alterations, with an HR of 0.56 (95% CI, 0.39–0.80; P = .001). Survival was similar between patients with gain alterations who did not receive targeted therapy and those without gain alterations (HR, 1.04 [95% CI, 0.88–1.23]; P = .612).
Because patients receiving matched therapy differed from untreated patients in factors including age, tumor type, and number of previous treatment lines, the investigators also performed an exploratory inverse probability of treatment-weighted analysis.
After adjustment, receipt of CGP-guided targeted therapy remained associated with longer overall survival, with an adjusted HR of 0.72 (95% CI, 0.58–0.89; P = .002). Among patients who received targeted treatment, median progression-free survival was 15.6 months. The objective response rate was 40.0%, including two complete responses and 20 partial responses, while the disease control rate was 74.5%.
What Do These Findings Mean?
The study provides a large real-world assessment of both the opportunities and practical limitations associated with CGP in GI oncology. CGP successfully generated results in approximately 90% of submitted samples and identified additional potentially actionable alterations in more than one-quarter of patients with advanced disease.
At the same time, only a minority ultimately received treatment based on these findings. Limited access to appropriate clinical trials or treatment programs and poor performance status were among the major barriers. The authors therefore emphasized two practical considerations: selecting appropriate tumor specimens and performing molecular profiling early in the advanced-disease course, while patients remain clinically able to receive additional therapies or participate in clinical trials.
Importantly, the reported survival findings should be interpreted in the context of the study’s retrospective, nonrandomized design. The cohort included heterogeneous GI tumor types, genomic alterations, treatment histories, and matched therapies. Although the IPTW analysis adjusted for measured clinical differences, the study cannot establish that CGP-guided therapy itself caused the observed improvement in survival.
The investigators also noted that the study was not designed to directly compare comprehensive profiling with smaller, tumor-specific gene panels.
Overall, the findings support the integration of CGP within comprehensive cancer centers, particularly where Molecular Tumor Boards, clinical trials, and mechanisms for accessing matched therapies are available. They also highlight that realizing the potential of precision oncology requires more than identifying a molecular target: testing must occur early enough, with adequate tissue and with realistic pathways for patients to access the therapies their genomic results identify.
The full article is available in JCO Precision Oncology.