Comment on “Soluble CD46 improves detection of hepatic steatosis by contrast-enhanced computed tomography before liver surgery”

Dear Editor, We read with great interest the recent work by Kupke et al1 entitled “Soluble CD46 improves detection of hepatic steatosis by contrast-enhanced computed tomography before liver surgery.” The authors addressed a clinically relevant question – preoperative identification of moderate-to-severe hepatic steatosis – and proposed integrating soluble CD46 (sCD46) with portal-venous-phase contrast-enhanced CT-based quantification. Compared with imaging metrics alone, the combined model provided superior discrimination of ≥grade 2 steatosis, while requiring no additional imaging sequence or radiation exposure, which enhances its feasibility for real-world implementation. Furthermore, the authors not only demonstrated statistical improvement but also supplied explicit cut-offs and a usable score, which facilitates external reproducibility – an uncommon but valuable feature in diagnostic research. Importantly, the main contribution of this work does not lie in introducing a “new biomarker” per se, but in structuring preoperative steatosis assessment through a “blood-marker + quantitative imaging” strategy, shifting the paradigm from qualitative impression to actionable quantification. This direction aligns better with how information is actually consumed within surgical decision-making. Several open questions remain. This was a single-center study with limited sample size, particularly for ≥grade 2 cases, and thus external validation is warranted to assess model stability and generalizability. As the model was not benchmarked against MRI-PDFF (proton density fat fraction) or chemical-shift imaging, the extent to which its incremental value is indispensable versus supplementary remains uncertain. Moreover, no perioperative hard outcomes (e.g. complications, postoperative liver failure, donor quality assessment) were evaluated; it is therefore unclear whether improved diagnostic granularity will translate into gains at the level of decision or prognosis. The authors also disclosed patents related to sCD46, which further underscores the need for validation in independent cohorts. In our view, three directions merit attention in future work: (1) external validation in steatotic liver disease populations with heterogeneous etiologies and comorbidities (e.g. hepatocellular carcinoma HCC-associated liver, potential donors, metabolic phenotypes) to assess robustness in complex settings; (2) head-to-head comparison with MRI-PDFF, metabolic composite scores, or high-dimensional imaging signatures to define the boundaries and positioning of its incremental value; and (3) prospective evaluation using continuous-risk probability rather than a single cut-off, to quantify its impact on actual preoperative pathways and clinical endpoints. If these lines of evidence are strengthened, the imaging-plus-blood architecture may advance from a diagnostic refinement toward a usable clinical tool. Overall, the study does not seek to redefine steatosis conceptually but provides a practicable approach to make existing information more clinically usable. Its contribution resides in translation, not reinvention, which is particularly meaningful in surgical contexts where decision utility – not merely detection – is the ultimate metric. We anticipate that with external and prospective validation, this structured integration of blood markers with routine imaging may evolve from a diagnostic enhancement into an instrument influencing preoperative management and risk communication, thereby clarifying the scope and tier of its clinical value2.