Tumor ratio of unsaturated to saturated sulfatide species is associated with disease-free survival in intrahepatic cholangiocarcinoma

**Background:** Cholangiocarcinoma (CCA) is a malignancy of the bile duct epithelium with poor prognosis, often diagnosed at an advanced stage. Sulfatides are sulfated glycosphingolipids that, within the liver, are uniquely present in bile ducts and have been implicated in various cancers. Their role in CCA had not been previously explored. This study aimed to characterize sulfatide abundance and composition in intrahepatic CCA (iCCA) tumor tissue and correlate these features with clinical outcomes.

**Methods:** Sulfatides were analyzed using matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI) in fresh-frozen tissue from 17 iCCA patients, 10 HCC patients, 10 CRLM patients, and 16 tumor-distal control samples (from HCC/CRLM patients). High-resolution imaging was performed on a 9.4T FT-ICR mass spectrometer at 20×20 µm pixel size. A total of 41–44 distinct sulfatide species were identified. Total sulfatide intensity and ratios of hydroxylated/non-hydroxylated, unsaturated/saturated, and odd/even-chain species were calculated per pixel within tumor regions or bile ducts. Survival analyses used Cox proportional hazard models and Kaplan-Meier curves with dynamic thresholding at the 50% quantile. Between-group differences were tested using non-parametric ANOVA with Dunn's post-hoc test.

**Key Results:** Sulfatides localized exclusively to bile ducts in control tissue and to tumor areas in iCCA, CRLM, and HCC. Median summed sulfatide intensity per pixel was similar in iCCA and control tissue (47878 vs. 31776, P=1.00). CRLM showed markedly elevated sulfatide levels: 18-fold vs. control (P=0.027) and 47-fold vs. HCC (P<0.001). HCC sulfatide levels were comparable to control and iCCA. The ratio of unsaturated to saturated sulfatides was not significantly different between groups overall, but within the iCCA group, a high ratio was significantly associated with shorter disease-free survival (median 10 vs. 20 months, P=0.002 by Kaplan-Meier). Overall survival showed a trend in Cox analysis (P=0.041) but was not significant by Kaplan-Meier (P=0.15). No correlations were found between sulfatide features and TNM stage or tumor grade. CRLM tumors showed remarkable homogeneity with upregulation of four hydroxylated species (ST-OH [34:1], [40:1], [41:1], [42:1]). HCC was distinguished by absence of the common bile duct sulfatide ST-OH [42:1] and relative abundance of odd-chain species, with a 3.8-fold higher odd/even-chain ratio vs. control (P=0.043).

**Clinical Implications:** This exploratory study identifies the sulfatide saturation ratio as a potential prognostic biomarker for early recurrence in iCCA, though validation in larger independent cohorts is needed. The distinct sulfatide profiles across tumor types (iCCA, HCC, CRLM) raise the possibility of using sulfatide signatures for differential diagnosis, particularly in challenging cases distinguishing HCC from iCCA. The presence of sulfatides specifically in tumor tissue (not tumor-distal parenchyma) suggests potential utility for intraoperative margin assessment using real-time mass spectrometry. Mechanistically, the association between unsaturated sulfatides and worse outcomes may reflect activity of desaturases like SCD1, which are therapeutic targets in other cancers. Limitations include small sample size (n=17 iCCA), short follow-up (2 years), selection bias toward resectable patients, and the exploratory nature of statistical findings.