liv12636-sup-0001-Suppmat.docWord document4090K

Fig. S1. Three- and two-reference small RNAs were used to build an internal reference (normalization factor) for target miRNA relative quantification in two of the patients' sets (see Methods for details).

Fig. S2. Identification of miR-425-3p as a predictive marker for sorafenib therapy.

Fig. S3. Kaplan–Meier survival curves of HCC patients treated with sorafenib according to clinical variables.

Fig. S4. Comparison of sensitivity and specificity for survival prediction by miRNA levels or clinical variables in HCC patients treated with sorafenib.

Fig. S5. miR-425-3p levels are not correlated with aetiology of liver disease. HCCs from the training and the validation cohorts were grouped according to the aetiology of liver disease and miR-425-3p expression was analysed (P = 0.92 by Kruskal–Wallis test).

Fig. S6. miR-425-3p and sorafenib modulation in liver cancer cell lines.

Table S1. Applied Biosystems identification code (ABI ID, Life Technologies Inc.) of the seven selected miRNAs analysed in HCC samples.

Table S2. Baseline patients' characteristics.

Table S3. Baseline patients' characteristics of the cohort of HCC patients that did not experienced sorafenib treatment (control set) (6, 18).

Table S4. miRNAs significantly associated to time to progression (TTP) in the training set of HCC patients treated with sorafenib.

Table S5. miRNAs significantly associated to progression free survival (PFS) endpoint in the training set of HCC patients treated with sorafenib.

Table S6. miRNAs significantly associated to overall survival (OS; timeframe: from therapy start to end of follow-up) in the training set of HCC patients treated with sorafenib.

Table S7. Statistical dependence between clinical or molecular (miR-425-3p) variables was analysed using the Spearman's rank correlation coefficient (ρ) in the validation cohort.

Table S8. The seven selected miRNAs were tested for differential expression in samples classes according to the indicated clinical variable in the two patients' cohorts.

Table S9. Bioinformatics analysis of potential interactions of miRNAs with sorafenib was performed using the web-based database Pharmaco miR ( and miR-425-3p and sorafenib as input of the search.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.