We read with interest the paper by Granito et al¹ reporting on the clinical and diagnostic significance of anti-filamentous actin antibodies (A-FAA) in autoimmune hepatitis type 1 (AIH-1). The authors found that A-FAA, measured by a new commercially available ELISA based on a modified cut-off of 30 instead of the manufacturer’s 20 arbitary units (AU), strictly correlates with the smooth muscle antibody glomerular/tubular (SMA-G/T) pattern,² also known as the microfilament pattern, mostly seen in patients with AIH-1.¹ Their findings further indicate F-actin as the predominant, if not the sole, target of AIH-1-specific SMA reactivity, a notion that has been questioned in the past because of the inconsistent results obtained by several actin-based assays.^3,4

We agree with Granito et al¹ that many laboratories are unfamiliar with the interpretation of the immunofluorescence patterns of AIH-specific SMA; such a problem arguably generates an urgent need for a reliable and observer-independent molecular-based SMA detection assay.⁵ New commercially available A-FAA ELISAs could be viable alternatives for routine laboratories, as the likelihood of a false-negative result is relatively low, but the possibility of reporting inaccurate results by immunofluorescence remains high.

We would like to raise a few points on the basis of our recent experience with the A-FAA ELISA.

At variance with Granito et al¹—and using the same ELISA and modified cut-off of 30 AU—we found A-FAA seropositivity in a considerable proportion of SMA-G/T-seronegative patients with AIH-1 (fig 1A). The increased sensitivity of the A-FAA ELISA comes to the cost of its lower specificity, as it detects A-FAA in a considerable number of SMA-seronegative pathological controls, especially in patients with primary biliary cirrhosis and chronic hepatitis C (fig 1B).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

 (A) Scatter plot representation of the correlation between the ELISA titres (arbitary units, AU) of the anti-filamentous actin antibodies (A-FAA; x axis) and the titres of smooth muscle antibody glomerular/tubular (SMA-G/T) pattern by indirect immunofluorescence (IFL; y axis). (B) ELISA titres for A-FAA in 75 patients with autoimmune hepatitis type 1 (AIH-1), 67 with primary biliary cirrhosis (PBC), 31 with primary sclerosing cholangitis (PSC), 70 with chronic hepatitis C virus (HCV) infection, 50 with hepatitis B virus (HBV) infection and 43 healthy controls. According to the manufacturer’s instructions, a test is strong positive when the ELISA titre is >30 AU; weak positive between 20 and 30 AU; and negative at <20 AU. In agreement with Granito et al,¹ the mean (standard deviation) optical density at 450(5) nm in our 50 controls corresponds to 29.6 AU.

We also found through inhibition studies that AIH-1-specific SMA-G/T reactivity is not always abolished by F-actin as a solid-phase competitor; only in one of the three patients with AIH-1 were we able to absorb out immunofluorescence SMA-G/T reactivity by 70% using F-actin as a solid-phase competitor^6,7 (in the other two patients, SMA-G/T reactivity was inhibited by 23% and 47%).

Our inhibition studies suggest that F-actin is a likely, but not the only, target of AIH-specific SMA reactivity.⁵ We agree with the recent consensus statement of the international autoimmune hepatitis group affirming that: (1) the basic technique of choice at present for the routine testing of SMA patterns relevant to AIH is indirect immunofluorescence on a rodent multiorgan (kidney, liver and stomach) substrate and (2) the remaining targets of the AIH-1-specific microfilament reactivity will require further identification.⁵ We also think that larger studies are needed to consider the performance of the newly established A-FAA ELISA in terms of specificity and sensitivity.