Study Results Show Multidisciplinary Tumor Boards Lead to Improved OS for HCC Patients

Multidisciplinary tumor boards led to higher utilization of guideline-recommended curative therapies, which was associated with improved overall survival (OS) for patients with early-stage hepatocellular carcinoma.

Marina Serper, MD, MS

Multidisciplinary tumor boards led to higher utilization of guideline-recommended curative therapies, which was associated with improved overall survival (OS) for patients with early-stage hepatocellular carcinoma (HCC), according to an observational study conducted by the Veterans Outcomes and Cost Associated with Liver Disease (VOCAL) research group that was presented at the 2016 AASLD Liver Meeting.

"Significant variability was noted for care delivery in HCC in a diverse, national cohort of Veterans," said lead investigator Marina Serper, MD, MS, Division of Gastroenterology and Hepatology, Hospital of the University of Pennsylvania. "Studies to improve multidisciplinary approaches for HCC in the community are needed to increase rates of curative therapy and improve clinical outcomes."

Improving care delivery in HCC involves many factors, with treatment variation and care management at the forefront. In general, the treatment paradigm for HCC is complex, as it depends equally on liver function and tumor staging. As a result, care in HCC is variable due to provider expertise and a range of available therapeutic options.

Ninety-nine percent of patients in the study were males, and 54% were white. The mean age was 62 years (SD ±8). Thirty-six percent of patients were within Milan criteria and 45% of patients had a Child Pugh score of B or higher. Thirty-four percent of patients had an alpha‑fetoprotein (AFP) blood value ≥200 ng/ml.

Patients had an average of 2 lesions (range, 1-4). Overall, 17.5% had microinvasions and 7.2% had metastatic cancer. The top etiologies for liver disease were hepatitis C virus (HCV) infection and alcohol (39%), HCV alone (22%), alcohol alone (15%), and both HCV and hepatitis B virus (HBV; 5%). The liver characteristics at baseline consisted of a Barcelona Clinic Liver Cancer Stage B (33%), A (30%), C (17%), D (14%), and 0 (6%).

The most common HCC therapies received in the frontline setting were sorafenib, transcatheter arterial chemoembolization, or no therapy. In the early stage, there were fewer patients with HCC with a BCLC stage between 0 and 2 receiving curative therapy (resection or ablation) than expected based on treatment guidelines. Additionally, sorafinib was utilized earlier than guidelines indicated.

Clinical factors that negatively impacted the frontline treatment decision included age at diagnosis (odds ratio [OR], 0.77;P<.001), MELD score (OR, 0.93;P<.001), ECOG performance status of 3 to 4 (OR, 0.40;P<.001), baseline AFP ≥200 ng/mL (OR, 0.80;P= .024), metastatic disease (OR, 0.71;P= .048), and macrovascular invasion (OR, 0.60;P<.001). Those within Milan criteria were more likely to receive active therapy (OR, 1.42;P= .003).

Delivery variables that positively affected the receipt of active frontline therapy included evaluation by ≥1 specialist (OR, 1.60;P= .005) and an academic center affiliation for the treating physician (OR, 1.97;P<.001). Treatment was also more likely for those who saw a medical oncologist (OR, 2.56;P<.001) and a surgeon (OR, 1.67;P= .001) within 30 days of diagnosis. Those who saw a gastroenterologist (OR, 0.56;P<.001) or a palliative care specialist (OR, 0.24;P<.001) were less likely to receive active therapy.

Additionally, Serper noted, “there were differences found between regions, and these data were analyzed. Regions will be presented in an upcoming manuscript.” Care delivery in HCC did vary significantly in this diverse national cohort of Veterans. Furthermore, Serper stated, “This cohort had a higher HCC mortality rate than the overall population.”

Factors associated with improved OS by multivariable analysis were liver transplantation (HR, 0.16), resection (HR, 0.33), ablative therapy (HR, 0.50), TACE (HR, 0.53), and sorafenib (HR, 0.76). Although care by a hepatologist was not associated with higher odds of receiving HCC therapy, seeing this specialist within 30 days of diagnosis did lead to improved OS (HR, 0.77;P<.001). Additionally, evaluation by a surgeon also improved OS chances (HR, 0.72).

Evaluation by ≥1 specialist, although it resulted in higher use of active therapy, did not result in an improvement in OS (HR, 1.13; 95% CI, 1.00-1.27;P= 0.52); however, evaluation by a multidisciplinary tumor board did positively influence survival (HR, 0.87; 95% CI, 0.81-0.94;P<.001).

"After adjusting for tumor staging and patient comorbidities, the type of managing HCC specialist, academic affiliation, and geographic location were independently associated with receipt of curative therapy," said Serper. "Care by hepatology was not associated with higher odds of HCC therapy, but was associated with improved survival and multidisciplinary tumor boards were associated with improved survival."

Several limitations of the study led the researchers to call for additional research on the impact of tumor boards, primary the fact that the study was retrospective in nature and that there was not a control arm. Additionally, the predominantly male population and lower access to transplantation may have impacted the results.


Serper M, Kaplan DE, Taddei TH, et al. Care Delivery affects Survival in Hepatocellular Carcinoma: A &lsquo;Real World&rsquo; Veteran Cohort Study. Presented at: AASLD Liver Meeting; Boston, Massachusetts, November 11-15, 2016. Abstract 170.

In the observational study, treatment factors were evaluated by characterizing the utilization of HCC therapies along with care management factors, such as patient, provider, and care delivery. All patients enrolled in the study were from the VA hospital system, and were selected from an administrative database from January 2008 to December 2010. All patients were followed until December 2014. After medical chart review, there was a confirmation of 6827 cases of HCC by ICD-9CM, with 5339 confirmed by chart review. Those treated outside of the VA were excluded (n = 1126), leaving 3998 patient records for review.