Molecular subtype and response to dasatinib, an Src/Abl small molecule kinase inhibitor, in hepatocellular carcinoma cell lines in vitro


  • Potential conflict of interest: Dr. Finn consults for and received grants from Bristol-Myers Squibb

  • View this article online at


Hepatocellular carcinoma (HCC) is the fifth most common malignancy and is the third leading cause of cancer death worldwide. Recently, the multitargeted kinase inhibitor sorafenib was shown to be the first systemic agent to improve survival in advanced HCC. Unlike other malignancies such as breast cancer, in which molecular subtypes have been clearly defined (i.e., luminal, HER2 amplified, basal, etc.) and tied to effective molecular therapeutics (hormone blockade and trastuzumab, respectively), in HCC this translational link does not exist. Molecular profiling studies of human HCC have identified unique molecular subtypes of the disease. We hypothesized that a panel of human HCC cell lines would maintain molecular characteristics of the clinical disease and could then be used as a model for novel therapeutics. Twenty human HCC cell lines were collected and RNA was analyzed using the Agilent microarray platform. Profiles from the cell lines in vitro recapitulate previously described subgroups from clinical material. Next, we evaluated whether molecular subgroup would have predictive value for response to the Src/Abl inhibitor dasatinib. The results demonstrate that sensitivity to dasatinib was associated with a progenitor subtype. Dasatinib was effective at inducing cell cycle arrest and apoptosis in “progenitor-like” cell lines but not in resistant lines. Conclusion: These findings suggest that cell line models maintain the molecular background of HCC and that subtype may be important for selecting patients for response to novel therapies. In addition, it highlights a potential role for Src family signaling in this progenitor subtype of HCC. (HEPATOLOGY 2013)

The need for progress in the treatment of hepatocellular carcinoma (HCC) has been highlighted by the rapid growth of the disease in the past decades.1, 2 In addition, at this time only one systemic agent, sorafenib, has been shown to be effective in treating the disease.3, 4 Historically, new systemic agents in liver cancer treatment have been evaluated irrespective of any patient or tumor-specific biology or predictive markers. Not surprisingly, many of these have not demonstrated significant clinical benefit, as they have approached HCC as one disease entity.5 We have since learned that patient selection is critical for the success of novel targeted agents in cancer medicine. For example, it was only after the completion of large negative clinical studies that mutations in the epidermal growth factor receptor (EGFR) were found to be associated with benefit to EGFR tyrosine kinase inhibitors in nonsmall-cell lung cancer.6, 7 Therefore, models that recapitulate human HCC may be of benefit in generating hypotheses that can then be validated in the clinic rather than continued empiric drug development using a “one-size-fits-all-approach.” In addition, cell lines originating from human tissue may more closely reflect clinical biology, rather than models engineered to reflect one specific alteration.

Gene expression profiling of human tissue has furthered our understanding of HCC and highlighted the molecular diversity of this disease.8-13 While we know that HCC most often develops in the setting of chronic liver disease, identification and validation of driving genetic alterations is still lacking. Laboratory models that recapitulate the molecular diversity of HCC would be of use to query the effectiveness of new targeted agents and potentially identify predictive markers of response to these agents.

Previous studies in breast cancer have shown that using a large panel of cell lines in vitro can recapitulate the molecular subgroups of the clinical disease in question.14, 15 In addition, these models have been used to generate hypotheses to then test prospectively in the clinic.14, 16, 17 In similar fashion, the clinical development of new therapeutics in HCC may benefit from such an approach.

We hypothesized that the described molecular subtypes in HCC clinical material would be maintained in a large enough panel of human-derived HCC cell lines. Further, to determine the potential importance for molecular subtype to predict for response to novel targeted therapies, we evaluated the antiproliferative effects of dasatinib, a small molecule tyrosine kinase inhibitor of the Src family kinases,18 in our molecularly characterized panel of cell lines.

The Src-family of tyrosine kinases (SFK) has nine members: Lyn, Fyn, Lck, Hck, Fgr, Blk, Yrk, Yes, and c-Src. Of these, c-SRC is the best studied and most frequently implicated in oncogenesis.19 c-SRC encodes a nonreceptor tyrosine kinase that, when activated, is involved in several pathways associated with oncogenesis including cellular proliferation, survival, migration, and angiogenesis.19 In HCC specifically, increased SRC activity has been described20-22 and in some studies has been correlated with an early stage phenotype.21

Building from the experiences in other solid tumors that preclinical models can represent the molecular heterogeneity of clinical disease, we tested this hypothesis in HCC. Specifically, we sought to demonstrate that there would be an association between the molecular subgroup of human HCC and response to the Src/Abl inhibitor dasatinib. Ultimately, the goal would be to identify potential biomarkers of response to dasatinib and to assist in patient selection and define a role for such an approach in HCC with molecularly targeted therapeutics in the future.

Materials and Methods

Cell Lines, Cell Culture, and Reagents.

The cell lines used in the analysis included SNU 449, SNU 475, SNU 423, SNU 387, SNU 182, PLC/PRF 5, HEP 3B, SK HEP 1, HEP G2, SNU 398, HLE, JHH4, JHH 6, HLF, HUH 7, JHH 5, HUH 1, JHH 2, JHH 7, JHH 1. SNU 449, SNU 475, SNU 423, SNU 387, SNU 182, PLC/PRF 5, HEP 3B, SK HEP 1, HEP G2, SNU 398 and were obtained from American Type Culture Collection (Rockville, MD). The cell lines HLE, JHH4, JHH 6, HLF, HUH 7, JHH 5, HUH 1, JHH 2, JHH 7, and JHH 1 were obtained from the Japanese Collection of Research Bioresources (Osaka, Japan). All cell lines were cultured in RPMI 1640 (Cellgro, Manassas, VA) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 2 mmol/L glutamine, and 1% PSF (Irvine Scientific, Santa Ana, CA).

Transcript Microarray Analyses.

Briefly, cells were grown to log phase and then RNA was extracted using the RNeasy Kit (Qiagen). The purified RNA was eluted in 30-60 μL DEPC water and the quantity of RNA measured by spectral analysis using the Nanodrop Spectrophotometer. RNA quality was determined by separation of the RNA by way of capillary electrophoresis using the Agilent 2000 Bioanalyzer. Microarray hybridizations of 20 HCC cell lines were performed using the Agilent Whole Human Genome 4x 44 K platform.

Characterizations of individual HCC cell line transcripts was performed by comparison to an HCC cell line mixed reference pool of RNA and were conducted on a single slide in which the cell line mixture RNA was labeled with cyanine-3 and RNA from the individual cell line with cyanine-5. The mixed reference complementary RNA (cRNA) pool consisted of equal amounts of cRNA from each of the HCC cell lines used in the study except JHH1, which was obtained at a later date. Microarray slides were read using an Agilent Scanner and Agilent Feature Extraction software v. 7.5 was used to calculate gene expression values. Data were normalized as described.14 Gene expression data analysis was subsequently conducted in R-project (build 2.11.1).

Data for clinical samples was obtained from the Gene Expression Omnibus (GEO) database (accession codes: human microarray platform, GPL1528; human HCC microarray data, GSE1898 and GSE4024).8 Data for the current study can be accessed at GSE35818.

Unsupervised Clustering of Gene Expression Data.

Expression data from 20 cell lines was clustered using an unsupervised hierarchical clustering protocol. To minimize random noise, genes with variances in the upper 25% quartile were selected. The distance matrix was calculated using the Pearson correlation and the histogram was generated using complete linkage clustering. Fisher's exact test was used to assess the relationship between response and subtype.

PAM Analysis.

Cross-dataset analysis was performed using the shrunken centroids technique outlined by Tibshirani et al.23 Human tumor data was obtained from previously published work of Lee et al.8 and included 139 human HCC samples (GSE1898). After removing transcripts with more than 50% missing data, 11,620 common transcripts were identified. Transcripts within each dataset were mean-centered and standardized to a variance of 1. Two classifiers were defined based on previously published work by Lee et al.,8 namely, the hepatoblast (HB) and the hepatocyte (HC) subtype. After the classifier was trained and cross-validated it was used to predict alternate class labels for the 20 cell lines in our dataset. The number of transcripts in the shrunken centroid classifier was chosen to minimize the false discovery rate from 20-fold cross-validation. The final analysis was done using a 300-gene classifier with a threshold of 3.0 (cross-validation error rate of 0%), although classification was unchanged using as few as 12 genes or as many as 10,000 genes.

Analysis of Differential Gene Expression.

For each of the two gene expression classes (HB-like and HC-like) derived from hierarchical clustering, we performed an analysis of the most significantly up-regulated and down-regulated genes among tumors. A Significance of Microarrays analysis was performed with 500 permutations of class labels to evaluate the significance of differentially expressed genes within each class. Up to 100 overexpressed transcripts and up to 100 underexpressed transcripts were selected from this analysis and are provided in the Supporting Tables 1 and 2.

Proliferation Assays.

Cells were seeded in duplicate at 5,000 to 10,000 cells per well in 24-well plates. The day after plating, dasatinib was added at 10 μM and 2-fold dilutions over six concentrations were performed to generate a dose-response curve. The number of dilutions was adjusted as necessary to capture the inhibitory concentration that reduced growth by 50% (IC50). Control wells without drug were also seeded. Cells were counted on Day 1 when drug was added as well as after 6 days when the experiment ended. After trypsinization cells were placed in Isotone solution and counted immediately using a Coulter Z2 particle counter (Beckman Coulter, Fullerton, CA). Viability was confirmed using a Coulter Vi-Cell counter (Beckman Coulter). Growth inhibition was calculated as a function of the number of generations inhibited in the presence of dasatinib versus the number of generations over the same time course in the absence of dasatinib.

Western Blot Analysis.

Cells in log-phase growth were treated with 100 nM of dasatinib and harvested at 30 minutes by washing in phosphate-buffered saline (PBS) and lysis at 4°C in RIPA lysis buffer. Insoluble material was cleared by centrifugation at 10,000g for 10 minutes and protein quantitated using BCA (Pierce Biochemicals, Rockford, IL). Protein content was resolved by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) electrophoresis, and transferred to nitrocellulose membranes (Invitrogen, Carlsbad, CA). Total Src expression was detected using a rabbit polyclonal antibody to the carboxy-terminus of human Src (Cell Signaling, Danvers, MA). Phopho-src was detected using rabbit polyclonal antibody to phospho-tyrosine-416 (EMD Biosciences, San Diego, CA). Blots were washed and incubated with a goat-antirabbit immunoglobulin G (IgG) horseradish peroxidase (HRP) conjugate (Upstate, Billerica, MA); developed using ECL Plus chemifluorescent reagent (Amersham Biosciences, Piscataway, NJ), and imaged using chemiflourescence. Densitometry was performed using the ImageJ 1.45s (NIH, Bethesda, MD) software. The densitometry results were presented as a relative percentage change from their corresponding controls.

Cell Cycle Analysis and Apoptosis Studies.

Effects of dasatinib on cell cycle were assessed using Nim-Dapi staining. Cells were plated evenly in control and experimental wells and allowed to grow to log-phase then treated with 100 nM dasatinib for 24 hours. To perform cell cycle analysis, cells were washed with PBS and trypsin was applied to release cells, which were then centrifuged at 3,000 rpm for 5 minutes. Supernatant was aspirated and cells were then resuspended in 100 μL of Nim-Dapi (NPE Systems, Pembroke Pines, FL) and gently vortexed. Cells were analyzed with UV using a Cell Lab Quanta SC flow cytometer (Beckman-Coulter). Apoptosis assays were performed using an Annexin V-FITC apoptosis detection kit (MBL, Woburn, MA) and flow cytometry. Cells were plated and treated as for cell cycle studies and exposed to 100 nM dasatinib for 5 days. After incubation, cells were processed as directed in the kit and analyzed using an FITC signal detector and propidium iodide (PI) detector using a Cell Lab Quanta SC flow cytometer. A one-sided t test was performed measuring significance of the increase in G0/G1 and the decrease in living cells in cell cycle and apoptosis assays, respectively.

Src Knockdown Using Lentivirus shRNA Gene Transduction.

Lentivirus transduction of a Src short hairpin RNA (shRNA) was used to knockdown Src expression. HLE and SNU 423 cells were plated in a six-well plate at a density of 2-2.5 × 105 cells/well and incubated overnight at 37°C and 5% CO2 in their respective media. Lentiviral particles containing c-SRC shRNA (Santa Cruz Biotechnology) was diluted with transduction media consisting of serum-free, antibiotic-free media and 5 μg/mL of Polybrene (Santa Cruz Biotechnology) to a multiplicity of infection (MOI) 7 and 13. Cells were washed with PBS and then 1 mL of diluted virus was added. Cells transfected with lentiviral particles containing scramble shRNA were used as a negative control and cop-GFP lentivirus was used as a control for transduction efficiency. After 18 hours, lentiviral shRNA media was removed and replaced with media containing serum and antibiotics. After overnight incubation the cells were trypsinized and placed in T-25 flasks with media containing 5 μg/mL puromycin (Santa Cruz Biotechnology). Cells were detached 72 hours posttransduction and analyzed by flow cytometry for green fluorescence protein (GFP) expression. All noninfected cells were killed by puromycin, and remaining cells had GFP expression, indicating 90%-100% transduction efficiency. Western blot was performed for total Src and phospho-Src as described above. Growth effects of shRNA were analyzed as described above comparing a vector control with the respective shRNA clone.


EGFR, epidermal growth factor receptor; GEO, Gene Expression Omnibus; HB, hepatoblast; HC, hepatocyte; HCC, hepatocellular carcinoma; PI, propidium iodide; SFK, Src-family of tyrosine kinases.


Human HCC Cell Lines Recapitulate the Molecular Heterogeneity of the Clinical Disease.

A total of 20 human HCC cell lines were used in the study. Available clinical data from the respective repositories are in Table 1. In an unsupervised clustering, the cell line panel breaks into two subgroups. This can be seen in Fig. 1. To validate our clustering results against previously published groupings in human disease, we trained shrunken centroid classifiers on a human expression dataset from Lee et al. Our classifiers showed 100% concordance with labels predicted by this external classifier, with these cell lines recapitulating the molecular subtyping described in human disease. Lee et al.24 initially described two large subgroups of HCC, Cluster A and Cluster B, that correlated with survival. However, in a follow-up study integrating data from rat fetal hepatoblasts and adult human hepatocytes with HCC from human and mouse models refined this classification into “HB” and “HC” groups which not only correlated with survival but also defined a molecular phenotype for these groups (i.e., “hepatoblast” versus “hepatocyte,” respectively). The cell lines therefore represent distinct subtypes of the clinical disease.

Figure 1.

Unsupervised cluster of 20 human HCC cell lines and sensitivity to dasatinib. Twenty human HCC cell lines were profiled using the Agilent platform as described. In an unsupervised fashion they break into two subgroups, both the HB and HC subgroups described be Lee et al.8 Cell lines are color-coded for their sensitivity to dasatinib in vitro.

Table 1. Twenty Human Liver Cancer Cell Lines
 Cell LineSourceSexHistologyReported Etiology
  1. Epidemiologic data available for the 20 cell lines used in the current study including their subtype classification, source, reported histology, sex, and associated hepatitis status. HCC, hepatocellular carcinoma; HBV, hepatitis B virus; HCV, hepatitis C virus.

3HLEJCRBMHCCNot reported
13SK HEP 1ATCCM“Hepatocarcinoma”Not reported
14HEP G2ATCCMHCCNot reported

Dasatinib Preferentially Inhibits Growth of “HB” Subtype of HCC Cell Lines In Vitro.

The 20 human HCC cell lines were evaluated for their sensitivity to the SRC/ABL tyrosine kinase inhibitor dasatinib. The calculated IC50 for each cell line and its molecular classification was determined (Table 2). There was a statistically significant correlation between molecular subtype and sensitivity to dasatinib (P < 0.01). The subtype most sensitive to growth inhibition by dasatinib was the HB subtype representing a “progenitor” subtype of HCC (Fig. 1). Using the subtype as classifier, only one cell line predicted to be resistant to dasatinib was actually sensitive (PLC-PRF5), and two cell lines predicted to be sensitive were actually resistant (JHH2 and SK Hep 1). This gives an overall specificity and sensitivity of subtype and association with positive response to dasatinib of 78% and 91%, respectively.

Table 2. Dasatinib Preferentially Inhibits “HB” Subtype of HCC
 Cell LineIC50 (nM)Molecular Subtype
  1. The IC50 values and molecular subgrouping of HCC cell lines is shown. Sensitivity to dasatinib is highly correlated with molecular classification (HB: hepatoblast subtype, HC: hepatocyte subtype, as defined by Lee et al.8).

1SNU 4493HB
2SNU 4759HB
4SNU 42314HB
5JHH 422HB
7SNU 38732HB
8SNU 18266HB
9PLC/ PRF 5118HC
10JHH 6152HB
12HEP 3B1427HC
13HUH 72000HC
14SK HEP 12546HB
15HEP G24000HC
16JHH 56000HC
17HUH 110000HC
18JHH 210000HB
19JHH 710000HC
20SNU 3989000HC

Identification of Differentially Expressed Genes and Sensitivity to Dasatinib.

To further determine a specific subset of genes that were predictive of response to dasatinib, an analysis of variance (ANOVA) identified 503 genes at a false discover rate (FDR) of <0.005 that were differentially expressed between dasatinib-sensitive and -resistant cell lines. Of interest, moesin (MSN), caveolin (CAV), and ephrin (EPH) family members (EPHRA) were up-regulated in the sensitive lines versus the resistant lines. All of these genes have been identified as being associated with dasatinib sensitivity in breast and lung cancer models, suggesting potential common molecular (not histological) determinates of dasatinib sensitivity.14, 25

Dasatinib Inhibits Src Phosphorylation Regardless of Sensitivity to Growth Inhibition.

Dasatinib is a multitargeted tyrosine kinase inhibitor. To evaluate the correlation between dasatinib's ability to block Src activity and its ability to inhibit proliferation in vitro, we performed western blots for phosphorylated src (pSrc) before and after dasatinib exposure. Figure 2 demonstrates that dasatinib is capable of blocking ppSRC at low nanomolar (nM) concentrations. The ability of dasatinib to block ppSRC is independent of its ability to inhibit growth. This is demonstrated by densitometry performed on the blot that indicates a significant decrease in ppSRC in both sensitive and resistant lines. While there is no significant decrease in the HEP G2 cell line that is resistant to dasatinib, the other resistant cell lines have similar decreases as the three sensitive lines tested. As expected, no change in total Src was seen during this time course.

Figure 2.

Dasatinib blocks phospho-Src regardless of its ability to block proliferation. Dasatinib significantly blocks phosphorylation of Src at tyrosine 416 in both cell lines that are sensitive to the antiproliferative effects of dasatinib and those that are resistant. In neither group does total Src change significantly with treatment. All cell lines were treated with 100 nM dasatinib for 30 minutes and western blots and densitometry were performed as described in the Materials and Methods.

Effects of Dasatinib on Cell Cycle and Apoptosis.

To better understand the mechanism by which dasatinib inhibits growth of HB-subtype cell lines, we evaluated dasatinib's effects on cell cycle and apoptosis in a subset of lines that were sensitive or resistant to the proliferation effects of dasatinib. For cell cycle, cells were exposed to dasatinib at 100 nM for 24 hours and then flow-cytometry using NimDAPI staining was performed. As can be seen in Fig. 3A, dasatinib effectively induces a G0/G1 arrest in cell lines that are sensitive to the compound in low nanomolar concentrations. This was not seen in cell lines resistant to dasatinib's growth inhibitory effects. Apoptotic effects were analyzed using Annexin V-FITC staining after a 5-day exposure of the same cell lines with 100 nM dasatinib (Fig. 3B). Similarly, an increase in apoptosis was seen after exposure to dasatinib in lines that had lower IC50 values than those that were higher and classified as resistant.

Figure 3.

Effects of Dasatinib on cell cycle and apoptosis. (A) Sensitive cell lines (IC50 <1,000 nM) show marked G0/G1 arrest and decrease in S-phase fraction as compared to resistant cell lines (IC50 >1,000 nM) after incubation with 100 nM dasatinib for 24 hours. (B) Similar observations were seen for the effects on apoptosis after 5 days of incubation using Annexin V, FITC, and PI staining. Living, FITC negative / PI negative; dead, FITC negative/ PI positive; early apoptosis, FITC positive / PI negative; late apoptosis, FITC positive/ PI positive. Solid bars, control, hatched bars, treated. Error bars represent standard error for at least two separate experiments and P-values represent the results of a one-sided t test measuring significance of the increase in G0/G1 and the decrease in living cells in cell cycle and apoptosis assays, respectively.

Knockdown of c-SRC by shRNA Does Not Affect Growth in Sensitive Lines.

Dasatinib is a multitargeted tyrosine kinase inhibitor of src and abl and other SFKs. To evaluate the effects of src knockdown on growth, we used a lentivirus to introduce an shRNA to c-src. Figure 4A demonstrates that in two cell lines that are sensitive to dasatinib, HLE and SNU 423, total and phospho-Src are decreased after transfection. Figure 4B shows that despite src knockdown, there is no effect on cell growth. This suggests that inhibition of src alone (versus other SFKs or other targets of dasatinib) by dasatinib may not explain its full activity in the dasatinib-sensitive cells.

Figure 4.

Src knockdown using lentivirus shRNA gene transduction. (A) An effective decrease in total src and phospho-src in two cell lines that are sensitive to dasatinib. In both HLE and SNU 423 cells, there is a significant decrease in the transfected clones (SRC7) as compared to the vector control cells (C7). A431 lysates were used as a positive control (+). (B) The results of cell count assays between the vector control and src knockdowns for both lines. As can be seen, despite knockdown of src and activated src, there are no effects on cell number over time.


Advances in the treatment of HCC have been limited by the lack of active agents. The lack of preclinical models in HCC has likely contributed to this limited success. Our aim in this work was to establish a panel of human HCC cell lines that recapitulate the previously described molecular subtypes in clinical cohorts and demonstrate that these subtypes may be important in linking targeted therapies with patient selection factors.

Initial work by Lee et al.24 described two large subgroups of HCC based on gene expression profiling, so-called “clusters A and B.” Further work from this group evaluated a signature derived from hepatic progenitor cells that identified a poor prognosis group that tightly clustered with rat fetal hepatoblasts and was named “HB.”8 The remaining group co-clustered with rat hepatocytes or was excluded from the hepatoblast core cluster and was associated with a better prognosis, the HC group.8 Overlaying the classification schemes, three distinct subgroups of HCC have been defined: those that have the best prognosis and originate from a more mature hepatocyte and are the HC subtype and overlay with cluster B (HCB), those with an intermediate prognosis that are cluster A and are the HC subtype (HCA), and those that have the worse prognosis and have a progenitor cell origin with the HB subtype (HB). We have shown that our panel recapitulates the two extremes of these groups, the HB group and the HC group. These observations are similar to an original report by Lee et al.26 that described differential gene expression of HCC cell lines in vitro. As has been done in breast cancer, here we determine that human cell lines in vitro can recapitulate the molecular heterogeneity of the clinical disease.14,15 Importantly, despite a fairly large number of cell lines, the HCA group is not represented. To that extent, observations made using cell lines do not encompass the full breadth of HCC and newer models are still needed.

In breast cancer, molecular subgroups have been linked to therapeutic interventions such as hormone directed therapy for the luminal subtype and HER2 targeted therapy for the HER2 subgrouping. In addition, using large panels of cell lines have led to preclinical observations linking subtype with new therapeutic interventions and have led to hypothesis-directed clinical research.14, 17, 27 In initiating this work, we hypothesized that given a large enough panel of human HCC lines, we would see a similar observation.

Src is ubiquitously expressed in human cancers and is associated with many aspects of transformation including proliferation, invasion, angiogenesis, and differentiation.28 In HCC, activation of Src has been implicated in the pathogenesis of the disease.21 Dasatinib, an orally active small molecule inhibitor of Src/ABL, was evaluated across our panel of cell lines. There was a strong correlation of sensitivity to dasatinib and cell lines representing the HB, progenitor subtype of HCC. This sensitivity was associated with induction of apoptosis and cell cycle arrest in sensitive lines. Src phosphorylation was blocked in both cell lines that were sensitive and resistant to the antiproliferative effects of dasatinib, suggesting measurement of this target alone and/or the effects of blocking the target would not be sufficient to select patients in the context of a clinical trial. Further, by knocking down src and activated src in cell lines sensitive to dasatinib, we did not observe any changes in cell proliferation. This suggest that blocking Src alone with dasatinib is not sufficient for its antiproliferative and proapoptotic effects. We can speculate that dasatinib's effects may be mediated through inhibition of other SFKs, abl, or other known and unknown targets of dasatinib in conjunction with src. This observation also highlights the potential importance of subtype dependence on dasatinib's effects on signaling in the progenitor (HB) subtype of HCC. Interestingly, a similar observation was made in human breast cancer models, pointing to a potential role for SFKs signaling in the nonluminal subtype of the disease, which also has been suggested to have progenitor origins.14

This is one of the first studies linking the molecular subtype of HCC with response to a novel therapeutic and generates a hypothesis that could be used for patient selection in the clinic. A study of dasatinib in advanced HCC (NCT00459108) was recently completed. While no biomarkers were used for patient selection, biopsy material from this study may be available to further validate this observation. Further, this panel of cell lines can be used in further pre-clinical studies to identify molecular subgroups of HCC that are more likely to respond to a given therapeutic and may help guide the development of new drugs in the treatment of this disease.


This work is supported by generous gifts from the Auerbach Family Gift for Emerging Therapies in Hepatocellular Carcinoma and the Pfleger Foundation (RSF).