Microvascular invasion in hepatitis B virus–related hepatocellular carcinoma: Another step toward preoperative evaluation?
Menée à partir de données portant sur 1 004 patients ayant subi une résection hépatique pour un carcinome hépatocellulaire lié au virus de l'hépatite B, cette étude évalue la performance d'un nomogramme, basé sur des critères clinico-pathologiques (diamètre tumoral, présence de plusieurs nodules, niveau sérique de l'alpha-fœtoprotéine supérieur à 20ng/ml, ...), pour estimer avant traitement le risque de micro-invasion vasculaire
The presence of microvascular invasion (MVI) is regarded as an important prognostic factor after resection for hepatocellular carcinoma.1,2 However, both the prognostic value and clinical relevance of MVI are being questioned.3,4 The main reason behind this disagreement lies in the fact that there is no accepted definition of MVI; the disagreement leads to difficulties in comparing the results from different studies. An associated problem is that the definition of MVI is not clearly stated in many publications. Regardless of the debate, the histopathologic presence of MVI does not help in our preoperative evaluation of the patient when selecting the optimal surgical strategy. We therefore read the work of Lei et al5 in this issue of JAMA Surgery with great interest. Previous attempts to create prognostic tools based on radiologic and clinicopathologic factors have failed to achieve broad acceptance.6 Although the limitations of retrospective, single-center studies are present, the work of Lei and colleagues5 provides us with results indicating that their nomogram might serve as an additional instrument for preoperative decision making. Furthermore, even if the statistical analysis behind the nomogram is extensive and meticulous, one concern remains: the specificity rates imply a substantial risk for both false-positive and false-negative predictions. Still, the nomogram is an interesting contribution to the challenging clinical evaluation of these patients. As the authors mention, we agree that it is necessary to evaluate the validity of the nomogram in a prospective, randomized clinical trial, preferably with several expert centers in collaboration.
JAMA Surgery , commentaire en libre accès, 2015