• Potential conflict of interest disclosure: Paul Calès has stock ownership in BioLiveScale Inc. that has a license for FibroMeters from Angers University.



We read with interest the comments by Sebastiani and Alberti on our new FibroMeter plus Fibroscan combination in chronic hepatitis C (CHC).

We acknowledge that SAFE was a significant improvement in the noninvasive diagnosis of liver fibrosis by increasing the diagnostic accuracy and reducing the rate of liver biopsy. SAFE uses the aspartate aminotransferase-to-platelet ratio index (APRI) as first-line test because it is easily calculated at bedside with virtually no cost. Moreover, Sebastiani and Alberti argue that APRI can be used as a first-line screening test.1 However, as stated in the pivotal study,2 SAFE for significant fibrosis (Metavir F≥2) requires liver biopsy in low APRI values due to an insufficient negative predictive value and does not require liver biopsy in high APRI values. Finally, SAFE does not use APRI as a screening test and has is inconvenient to select low-risk patients for an invasive procedure like liver biopsy. In fact, fibrosis evaluation in CHC does not correspond to a screening but to a diagnostic procedure for patient management and must use the most accurate tests. In this setting, FibroMeter and Fibroscan have been independently validated as accurate fibrosis tests in large series,3, 4 and their combination is the most accurate among six noninvasive fibrosis tests.5

We agree that significant fibrosis and cirrhosis are important diagnostic targets for, respectively, antiviral therapy and screening procedures. However, in clinical practice physicians have to successively use SAFE for F≥2 and then, in case of noninvasive “F≥2” diagnosis, SAFE for F4 to secondarily discriminate F4 from F2/3 patients. Our results clearly show that this successive use of algorithms devoted for a binary diagnosis of fibrosis significantly decreases the diagnostic accuracy and increases the rate of liver biopsy. Our new FM+FS classification circumvents this limitation by giving a precise evaluation (six diagnostic classes) of liver fibrosis in a one-step procedure.

We agree that F2±1 and F3±1 diagnoses in our classification might induce difficult decisions in clinical practice. A classical solution is to perform liver biopsy in these cases. FM+FS classification would thus require 36.8% biopsies, which is still significantly lower than with Successive SAFE (63.8%, P < 0.001) or Successive Bordeaux Algorithms (49.8%, P < 0.001). A modern solution is to evaluate the prognostic value of FM+FS classification for clinical events. Recent works have shown that noninvasive fibrosis tests had a prognostic value equal or superior to histological fibrosis staging6, 7 and that the FM+FS combination might increase this prognostic value.8 This clearly supports a possible patient management without liver biopsy.

In conclusion, a combination of noninvasive fibrosis tests is an excellent diagnostic tool for liver fibrosis in CHC patients. The fibrosis tests must provide the most accurate combination. Combining a blood test with a physical test such as Fibroscan seems the better clinically available choice. In this setting, the synchronous combination of FibroMeter with Fibroscan allows for a precise and accurate diagnosis of liver fibrosis, with a minimum or, even more, without any liver biopsies according to the physician choice.

Jérôme Boursier*, Paul Calès*, * Hepato-gastroenterology Department, CHU, Angers, HIFIH Laboratory, IFR 132, University of Angers, PRES UNAM, Angers, France.