A meta-analysis of palliative treatment of pancreatic cancer with high intensity focused ultrasound

Abstract

Background: Pancreatic adenocarcinoma is currently the fourth-leading cause of cancer-related death. Up to 60-90% of patients with advanced disease suffer cancer-related pain, severely impacting their quality of life. Current management involves primarily pharmacotherapy with opioid narcotics and celiac plexus neurolysis; unfortunately, both approaches offer transient relief and cause undesired side-effects. High intensity focused ultrasound (HIFU) is a non-invasive thermal ablation technique that has been used to treat pancreatic cancer. This meta-analysis aims to evaluate the role of HIFU in pain palliation of advanced unresectable pancreatic adenocarcinoma.

Methods: An electronic search was performed in PubMed Medline database up to the end of July 2016, for unresectable pancreatic cancer pain palliation with HIFU. Pertinent studies were identified through the PubMed search engine using the following keywords: HIFU, pancreas, pancreatic cancer, pain and palliation. Additional studies were included after manual search of the selected bibliographies. Pain palliation results reported in each study were analyzed using a logit-transformed random-effects model using the inverse variance method, with the DerSimonian-Laird estimator for τ², and Cochran's Q test for heterogeneity among studies. The I² was calculated to assess the percentage of the total variability in the different effect size estimates that can be attributed to heterogeneity among the true effects. A rank correlation test of funnel plot asymmetry was done to assess possible publication bias.

Results: The meta-analysis includes a total number of 23 studies with 865 patients, 729 with pancreatic cancer. The population enrolled ranges from 3 patients in the smallest series, up to 61 in the largest study. τ² (variance among studies) was 0.195, and I² (percentage of variation among studies) was 40% (95% CI: 1-64%); the Q test p-value was 0.026, indicating significant heterogeneity among studies. Among 639 patients treated with HIFU, 567 complained of pancreatic pain before the treatment and 459 patients experienced partial or complete pain relief after treatment. The random effects estimate of the proportion of patients with pain reduction was 0.81 (95% CI: 0.76-86).

Conclusions: HIFU appears to be an effective tool for pain palliation in advanced pancreatic cancer. Studies assessing treatment in patients with pancreatic adenocarcinoma are limited by factors such as small sample sizes and heterogeneity in clinical definitions and assessments. Prospective randomized and standardized studies are necessary to confirm the effectiveness of HIFU in relieving pain, and to evaluate for any potential impact on tumor control and patient survival.

Keywords: HIFU; High intensity focused ultrasound; Meta-analysis; Non-invasive treatment; Pain palliation; Pain relief; Pancreatic cancer.

Fig. 1

PRISMA Flowchart. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-analyses

Fig. 2

Random Effects Model – Studies included in the analysis. Proportion of patients with pain reduction

Fig. 3

Funnel plot demonstrating possible but not statistically significant publication bias in assessment of pain (P > 0.05). -Dashed diagonal lines indicate 95% CI