Molecularly Tailored Therapy in Advanced Pancreatic Cancer

TAILOR-PANC (DPCG-02) is an investigator-initiated, multicenter, randomized phase 2 trial evaluating whether molecularly tailored second-line therapy improves outcomes compared with standard-of-care treatment in patients with advanced pancreatic cancer.

Although most pancreatic cancers harbor KRAS mutations, a smaller proportion contain potentially actionable molecular alterations, including selected mutations, gene fusions, amplifications, or biomarkers such as mismatch repair deficiency. Some of these alterations can be targeted by medicines that are already available in Denmark, although the medicines may have been developed or approved primarily for other cancer types. Observational studies suggest that patients with pancreatic cancer who receive treatment matched to an actionable alteration may have better outcomes than patients who receive non-matched treatment. However, this strategy has not been adequately evaluated in a randomized setting.

Molecular profiling is performed before study allocation using tumor tissue and/or blood. Whole-genome sequencing is preferred, but targeted next-generation sequencing or whole-exome sequencing may be used when whole-genome sequencing is unavailable. The results are summarized in a personalized molecular report. A national multidisciplinary molecular tumor board, including pancreatic cancer oncologists, molecular biologists, pathologists, and computational biologists, reviews the molecular findings and assesses whether an alteration is clinically actionable. A molecular alteration is considered actionable when it can be matched to a specific anticancer treatment supported by clinical evidence and the treatment is available and reimbursed in Denmark.

Participants whose cancers harbor an actionable, reimbursed molecular alteration are randomized in a 1:1 ratio to:

Arm 1: molecularly tailored therapy selected according to the recommendation of the national molecular tumor board; or Arm 2: standard second-line therapy according to Danish clinical guidelines.

Because the molecular alteration determines the matched treatment, participants assigned to Arm 1 may receive different targeted treatments. The relevant drug-specific requirements, safety monitoring, dose modifications, and discontinuation criteria will apply to each treatment.

Participants whose cancers do not harbor an actionable, reimbursed molecular alteration are assigned to Arm 3, a non-randomized observational cohort receiving standard-of-care therapy according to Danish guidelines. Data from this cohort will support exploratory comparisons and provide information about outcomes among genomically profiled patients without an available matched treatment. Comparisons involving Arm 3 will be interpreted as non-randomized because molecular characteristics and other prognostic factors may differ between the groups.

The primary randomized comparison is progression-free survival between Arms 1 and 2. Approximately 68 participants with actionable alterations are required for randomization, with approximately 65 progression-free survival events planned for the primary analysis. The study is designed to provide approximately 80% power at a two-sided significance level of 0.05 to detect a hazard ratio of 0.50. This corresponds to the design assumption of an increase in median progression-free survival from 3 months with standard therapy to 6 months with molecularly tailored therapy. These values are statistical assumptions and do not represent guaranteed treatment outcomes.

Because actionable alterations are expected in only approximately 5%-10% of patients with pancreatic cancer, up to 1,200 genomically profiled participants may be enrolled to identify the required number for the randomized comparison. Recruitment is expected to take approximately 4 years, with approximately 1 additional year of minimum follow-up for the primary analysis.

The study will also characterize overall survival, tumor response, duration of response, safety, and patient-reported quality of life. Optional translational research will be conducted through the BIOPAC project under separate consent. Blood and available tumor samples may be used to explore molecular and circulating biomarkers associated with treatment response, resistance, and toxicity. Clinical and genomic data may also be used in exploratory artificial intelligence and machine-learning analyses intended to develop predictive models and identify potential therapeutic targets. These biomarker and computational analyses are exploratory and are not used as validated clinical tests within this study.