Efficacy of systemic oncological treatments in patients with advanced pancreatic cancer at high risk of dying in the short or medium-term: overview of systematic reviews

Patients with advanced pancreatic cancer (PC) have a high risk of dying in the short or medium-term. This overview aimed to assess the evidence regarding systemic oncological treatments (SOT) versus supportive care for advanced PC. We searched for systematic reviews (SRs) in MEDLINE, Embase, The Cochrane Library, Epistemonikos, and PROSPERO. Two authors assessed eligibility independently. Data extraction and methodological quality assessment were conducted by one author and cross-checked by another one. We evaluated the overlap of primary studies, performed a de novo meta-analysis, and assessed the certainty of evidence. Primary outcomes were overall survival (OS), quality of life (QoL), functional status (FS), and toxicity. We identified three SRs that assessed SOT versus supportive care in patients with advanced PC. All SRs had critically low methodological quality. At 12 months, OS improved with chemotherapy, radiotherapy followed by chemotherapy, and immunotherapy, but the certainty of the evidence supporting these findings is very low. The evidence on chemotherapy is very uncertain about its effects on QoL; it suggests a slight increase in toxicity and little to no difference in FS. The evidence on immunotherapy is very uncertain about its effects in toxicity. The identified evidence is very uncertain about the benefits of oncological treatments on OS and QoL in patients with advanced PC with a high risk of dying in the short or medium-term, so its use should be proposed only to selected patients. Further studies that include a thorough assessment of patient-centred outcomes are needed. •Overview of systematic reviews of treatments for advanced pancreatic cancer.•Most patients in an end-of-life setting are treated with oncological treatments.•Uncertainty about the benefits of oncological treatments for very advanced disease.•Few studies found and most do not include patient-centred outcomes.•Need for studies that specifically assess patient-centred outcomes.