NGS Testing of HCC Reveals Potential Biomarkers and Mutations for Treatment Matching

March 29, 2019
Lisa Astor

An analysis of the genotyping of tumor tissue samples from patients with hepatocellular carcinoma identified frequent alterations and potentially actionable mutations in cases of HCC, and helped to confirm the clinical utility of next-generation sequencing testing for matching patients to targeted therapies and immunotherapies.

An analysis of the genotyping of tumor tissue samples from patients with hepatocellular carcinoma (HCC) identified frequent alterations and potentially actionable mutations in cases of HCC, and helped to confirm the clinical utility of next-generation sequencing (NGS) testing for matching patients to targeted therapies and immunotherapies.

The study authors, led by James J. Harding, MD, of Memorial Sloan Kettering Cancer Center (MSK), explained in their report published inClinical Cancer Researchthat NGS could help identify patients likely to benefit from standard systemic therapies. “…Our data represent the first attempt to link real-time NGS to clinical practice and provide needed insight into feasibility and the potential impact of this investigational modality to direct patient care.”

Investigators completed prospective NGS testing of 134 samples from 127 unique patients who had been treated at MSK and compared the genomic findings to those from The Cancer Genome Atlas (TCGA).

“We confirm genomic findings derived from retrospectively analyzed cohorts of surgically-resected, early stage, liver-limited HCCs and extend these observations to an advanced patient population more representative of those treated as part of routine clinical practice,” the study authors wrote.

DNA from the tumor tissue samples and patient-matched normal blood were all tested with MSK-IMPACT, a hybridization capture-based NGS assay that has been authorized by the New York State Department of Health and the FDA. Testing was performed in a Clinical Laboratory Improvement Amendments (CLIA)—certified laboratory.

The assay tested 341-plus cancer-associated genes, and identified single-nucleotide variants, small indels, copy number alterations, and selected structural rearrangements in all exons and select introns. MSK-IMPACT was expanded during the study period to include 410 genes, which was tested in 70 cases, and later to 468 genes, which was used for 36 cases. Tumor mutational burden (TMB) and microsatellite instability (MSI) were also tested with the assay. TMB was calculated as the total number of nonsynonymous mutations divided by the number of bases analyzed, and MSI was defined by the percentage of unstable microsatellites of all tested microsatellites.

All identified alterations were classified as actionable or not on an evidence scale of 1 to 4. Level 1-2A alterations would indicate a standard therapeutic intervention; levels 2B-4 would indicate an investigational alteration and could direct a patient toward a relevant clinical trial.

Genomic Findings

Patients in the TCGA cohort had surgically resected liver-limited tumors whereas those in the MSK group also included those who were ineligible for curative resection or transplantation and those with locally recurrent/extrahepatic tumors. The median age of patients in the MSK cohort was 65 years (range, 15-85) and 79% were male. Forty-nine percent had a nonviral etiology while 17% and 35% had hepatitis B and hepatitis C virus, respectively. Forty percent had Barcelona Clinic Liver Cancer (BCLC) stage B disease.

Genomic alterations were found in 95% (n = 127) of the samples, with the most frequently mutated genes (≥10%) beingTERT(55.6%),CTNNB1(35.7%),TP53(32.5%),ARID1A(12.7%),andTSC2(10.6%). Other alterations in canonical members of the WNT pathway beyondCTNNB1were also identified.

“Our data are consistent with prior reports regarding the frequency and overall distribution of oncogenic alterations in HCC. Specifically, we confirm that mutations predicted to result in WNT/β-catenin activation are present in nearly half of all HCCs, and represent a unique biological subset,” Harding et al wrote.

Copy number alterations included amplifications inCCND1/FGF19,MYC,MET,VEGFA, and loss inHLA-B. Fusion events were not common in the cohort.

The investigators rarely foundTP53mutations in conjunction with WNT pathway alterations (P= .003), which helped to separate 2 genomic subsets of patients with HCC.TP53mutations were more commonly identified in virally mediated HCC versus nonviral disease (45% vs 21%, respectively;P= .001), whereas WNT pathway alterations were less common in virally mediated disease (41% vs 60%;P= .7), although this was not found to be statistically significant.

The median TMB among the MSK group was 4.08 and demonstrated a lower variance in comparison with other solid histologies (standard derivation, 2.71). No MSI-high or hypermutation was noted in the cohort.

Even with the differences in the patient characteristics between the 2 groups, there was no significant difference found in the relative proportion of altered genes. Clinicopathological factors, such as age, ethnicity, and histologic grade, were not found to be associated with genomic alterations.

Findings Predictive of Response to Therapy

Targeted Therapy

Eighty-seven patients (68.5%) were treated with systemic therapy, and the other 38 (29.9%) were not eligible for systemic treatments. Sorafenib (Nexavar) treatment was given to 81 patients (63.8%), and the median progression-free survival (PFS) among these patients was 4.8 months. The median overall survival (OS) with sorafenib was 16.4 months, which compared favorably with findings from large published trials.

The investigators looked for potential genomic biomarkers that could be predictive of response to sorafenib. For example, activation of the mTOR pathway has previously been associated with resistance to sorafenib and worse outcomes. Patients who had activating mutations in the PI3K/mTOR pathway that were treated with sorafenib had lower rates of clinical benefit (8.3% vs 40.2%;P= .05), a shorter median PFS (1.9 vs 5.3 months; HR, 3.8; 95% CI, 2.0-7.5;P<.0001), and a shorter median OS (10.4 vs 17.9 months; HR, 2.5; 95% CI, 1.21-5.31;P= .01) than those who did not have PI3K/mTOR pathway mutations.

Alterations in receptor tyrosine kinases that may be targeted in treatment with sorafenib did not have any effect on outcomes. Activating mutations in the WNT/MAPK and TP53 pathways had no effect on outcomes either.

VEGFAamplification has been identified as a potential biomarker for patients who respond very well to sorafenib treatment, yet these alterations were only observed in 3.9% of patients. Additionally, there were no associations with clinical benefit noted anecdotally.


Thirty-one patients (24.4%) received treatment with an immune checkpoint inhibitor as systemic therapy. A majority received anti—PD-1/PD-L1 monotherapy, 5 received PD-1 inhibition in combination with a targeted antibody directed at CTLA-4, KIR, or LAG-3.

Of the 27 evaluable patients, 1 had a complete response and there were 2 partial responses for an objective response rate of 11.1%. An additional 10 patients had stable disease. The median PFS was 5.4 months and the median OS was 12.9 months.

Radiographic response to immune checkpoint inhibition was correlated with activating mutations in the WNT pathway, yet no benefit was seen in the 10 patients identified with WNT pathway alterations. These patients had a shorter median PFS (2.0 vs 7.4 months; HR, 9.2; 95% CI, 2.9-28.8;P<.0001) and a numerically shorter OS (9.1 vs 15.2 months; HR, 2.6; 95% CI, 0.76-8.7;P= .11) compared with those without such alterations. No other pathways were correlated with responsiveness or resistance to checkpoint inhibitor therapy, the study authors noted.

TMB also did not seem to have an impact on response to treatment as the median TMB was low for both responders and nonresponders (3.8 and 3.9, respectively;P= .98) and the median PFS was similar above and below the median TMB level of 3.8.

Treatment Matching

At least 1 actionable mutation was found in 24% of patients, none of which were level 1 or 2A alterations.

Among the potentially actionable mutations found were truncating and inactivating mutations in mTOR modulators,TSC1(<1%) andTSC2(7%), both of which were classified as level 2B alterations, as well asMETamplifications (1.5%) andBRCA2alterations (<1%). Truncating or homozygous deletions inPTENwere found in 3.9% of patients, which was categorized as a level 4 alteration.FGF19amplifications (6.3%),NOTCH1alterations (1.5%), and oncogenic missense mutations inHRAS(<1%) were all also considered level 4 in terms of their actionability.

Of 87 patients that received systemic therapy, 5 were matched to agents targeting mTOR or MET, including everolimus (Afinitor) for 4 patients withTSC1/2alterations and a VEGFR/MET antibody for a patient withMETamplification. Sorafenib was ineffective in the frontline for 4 of these 5 patients who experienced early progression; second-line targeted therapy was more beneficial for each patient leading to stable disease. The fifth patient received everolimus as first-line therapy for 5 months before progressing.

“Although the direct clinical impact of prospective NGS for HCC clearly has less utility than in other solid tumors, such as lung cancer and melanoma, 24% of tumors had potentially actionable alterations that could be targets for currently available FDA-approved drugs, or agents in active clinical development,” the investigators wrote. “As an example, our data and other series specifically support continued investigation of mTOR inhibitors in tumors withTSC1/2alterations.”

The authors noted, however, that tumor heterogeneity was not fully captured in this analysis and that they were also limited by the small sample size. They also did not address such confounding variables as etiology.


Harding JJ, Nandakumar S, Armenia J, et al. Prospective genotyping of hepatocellular carcinoma: clinical implications of next-generation sequencing for matching patients to targeted and immune therapies [published online February 6, 2019].Clin Cancer Res.doi: 10.1158/1078-0432.CCR-18-2293.