Total cellular RNA from 10-um thick sections was extracted with a proteolytic enzymatic system after deparaffinization. Two proteinase K digestion periods were compared to assess the quality of the extracted RNA from FFPE liver tissue of up to 10 years old: a 5 hours at 60šC system, previously reported by Edamoto et al (1996) and a modified 4 days at 42šC method. Twenty-seven liver specimens from individual patients (16 from serum HCV-RNA positive patients and 11 from serum HCV-RNA negative patients), who underwent orthotopic liver transplantation from 1988 to 1996, were analyzed. Reverse transcription (RT) was performed with 1 ug of total RNA and random primers. HCV-RNA was amplified using primers derived from the highly conserved 5' noncoding region of the HCV genome, as previously described by Sánchez-Tapias et al (1996). The reaction was performed by using an initial 5-minute step at 94šC, followed by 35 cycles of 1 minute at 94šC, 1 minute at 55šC, and 1.5 minutes at 72šC, with a final 10-minute completion step at 72šC. Nested PCR was carried out using 3 uL of the RT-PCR product in a final volume of 50 uL with inner primers NCR3 and NCR4 (Garson et al, 1990) or NCR3 (Garson et al, 1990) and p126 (Lim et al, 1994). PCR was carried out under the same conditions described above. Albumin mRNA was always amplified as an internal control for nucleic acid extraction and reverse transcription steps. Albumin mRNA amplicon was only visible when reverse transcription of RNA was previously carried out (data not shown), confirming absence of pseudogene amplification. The expected size of the HCV amplicons was 171 bp or 286 bp, depending on the primers used. No false HCV-RNA positive results were detected. HCV-RNA in serum was amplified as previously described by Sánchez-Tapias et al (1996). Data were analyzed statistically using the McNemar test or binomial exact probability when necessary.

As it is shown in Table1, a higher sensitivity was always achieved with the extensive method (42šC during 4 days), regardless of the amplicon size used: 15 of 16 versus 4 of 16 with 286 bp product (p < 0.005) and 16 of 16 versus 11 of 16 with the 171 bp product (p = 0.06). This could be caused by a higher degradation of HCV-RNA when the shorter method (60šC during 5 hours) was used. Only 4 of 16 and 11 of 16 (p < 0.02) samples were amplified when the 286 bp and 171 bp fragments were tested. In contrast, no significant difference was observed when the two fragments were assayed with the longer (42šC during 4 days) method (15 of 16 versus 16 of 16).

In the present study, the sensitivity observed with the shorter method (60šC during 5 hours) was similar to that previously described by Edamoto et al (1996). However, we have shown that it can be improved to 100% when a prolonged digestion and an appropriate target length are used. Our data also show that, for amplified targets of 150-300 nt, the effect of prolonged tissue storage on up to 10-year-old FFPE blocks does not seem to affect the results greatly.

Once the optimal conditions were established, they were applied to a second group of 18 FFPE archival trucut liver biopsies obtained from 1971 to 1985 (mean 18.4 -\+ 3.8 years). At the time of sampling, 10 of the 18 FFPE biopsies were from subjects diagnosed with NANB hepatitis, and eight were from patients with nonhepatic diseases. The length of the biopsy specimens selected to extract enough RNA to be amplified was at least 7 mm.

Fourteen of the paraffin blocks were rebuilt to be adapted to modern microtomes. HCV-RNA detection was achieved in 7 of 10 (70%) samples from patients with chronic NANB hepatitis, including three that were stored for more than 20 years (Fig. 1). The lack of HCV-RNA detection in three patients with chronic hepatitis, that have been recently serologically confirmed as HCV-RNA positive, may be caused by several factors: a) as mentioned above, the older original blocks had to be rebuilt, and this second paraffin inclusion could have affected the fragile integrity of HCV-RNA; and b) the aging effect is more important in more than 15-year-old FFPE tissue.

In conclusion, we have shown that the methodology developed in the present report is suitable for retrospective studies in FFPE archival liver biopsies, allowing, for the first time, successful HCV-RNA amplification from biopsies stored for more than 20 years. Therefore, this technique could be useful in the study of the natural history of HCV infection. Furthermore, because, in many instances, serum samples were not collected at the time of liver biopsy, this methodology can be useful for retrospective diagnosis of HCV infection.

References

Edamoto Y, Tani M, Kurata M, and Abe K (1996). Hepatitis C and B virus infections in hepatocellular carcinoma. Analysis of direct detection of viral genome in paraffin embedded tissues. Cancer 77:1787-1791.

Garson JA, Ring C, Tuke P, and Tedder RS (1990). Enhanced detection by PCR of hepatitis C virus RNA. Lancet 336:878-879. [Letter].

Lim HL, Lau GKK, Davis GL, Dolson DJ, and Lau JYN (1994). Cholestatic hepatitis leading to hepatic failure in a patient with organ-transmitted hepatitis C virus infection. Gastroenterol 106:248-251.

Mies C (1994). Molecular biological analysis of paraffin-embedded tissues. Hum Pathol 25:555-560.

Sanchez-Tapias JM, Forns X, Ampurdan|$$|Aaes S, Tito L, Planas R, Viver JM, Acero D, Torres M, Mas P, Morillas R, Forne M, Espinos J, Llovet JM, Costa J, Olmedo E, Lopez-Labrador FX, Jimenez de Anta MT, and Rodes J (1996). Low-dose alpha-interferon therapy can be effective in chronic hepatitis C. Results of a multicentre, randomised trial. Gut 38:603-609.