Background:
We performed a critical overview of published systematic reviews (SRs) of chemotherapy for
advanced and locally advanced pancreatic cancer, and evaluated their quality using AMSTAR2 and ROBIS tools.
Materials and Methods:
PubMed and Cochrane Central Library were searched for SRs on 13th June 2020. SRs with metaanalysis which included only randomized controlled trials and that had assessed chemotherapy as one of the treatment
arms were included. The outcome measures, which were looked into, were progression-free survival (PFS), overall survival (OS), and adverse events (AEs) of grade 3 or above. Two reviewers independently assessed all the SRs with both
ROBIS and AMSTAR2.
Results:
Out of the 1,879 identified records, 26 SRs were included for the overview. Most SRs had concluded that gemcitabine-based combination regimes, prolonged OS and PFS, but increased the incidence of grade 3-4 toxicities, when
compared to gemcitabine monotherapy, but survival benefits were not consistent when gemcitabine was combined with
molecular targeted agents. As per ROBIS, 24/26 SRs had high risk of bias, with only 1/26 SR having low risk of bias. As
per AMSTAR2, 25/26 SRs had critically low, and 1/26 SR had low, confidence in the results. The study which scored
‘low’ risk of bias in ROBIS scored ‘low confidence in results’ in AMSTAR2. The inter-rater reliability for scoring the
overall confidence in the SRs with AMSTAR2 and the overall domain in ROBIS was substantial; ROBIS: kappa=0.785,
SEM=0.207, p<0.001; AMSTAR2: kappa=0.649, SEM=0.323, p<0.001.
Conclusion:
Gemcitabine-based combination regimens can prolong OS and PFS but also worsen AEs when compared to
gemcitabine monotherapy. The included SRs have an overall low methodological quality and high risk of bias as per AMSTAR2 and ROBIS respectively.
Optimising treatment pathways for borderline and locally advanced pancreatic cancer: an adaptive personalised approach
