Detection of hepatitis C virus RNA by reverse-transcriptase and polymerase chain reaction: clinical applications of quantitative analysis

Clinical and Diagnostic Virology, 1 (1994) 289-297 © 1994 Elsevier Science B.V. All rights reserved 0928-0197/94/$07.00 DIAVIR 00031

Clinical and Diagnostic Virology

Detection of hepatitis C virus RNA by reversetranscriptase and polymerase chain reaction: clinical applications of quantitative analysis Maria Lorena Abate, Paola Manzini, Francesco Negro, Maurizio Baldi, Giorgio Saracco, Paolo Piantino, Maurizia Rossana Brunetto and Ferruccio Bonino* Laboratory of Department of Gastroenterology, Molinette Hospital, Corso Bramante 88, 10126 Turin, It'aly (Received 11 May 1993;revised 26 August 1993;accepted 26 August 1993)

Abstract Polymerase chain reaction (PCR) applications to diagnostics allowed the detection of viral nucleic acids in expected and unexpected clinical circumstances. This has raised some scepticism on the practical usefulness of PCR in the routine laboratory and emphasized the need for quantitative analysis. We addressed this question detecting HCV-RNA by a single step RT-PCR in serum samples from 50 patients with chronic non-A, non-B hepatitis included in clinical trials for recombinant alphainterferon therapy. We obtained at least 5 serum specimens from each patient (baseline, during and after therapy samples) during an 18-month mean follow-up (range 12-45 months). RT-PCR was performed on total RNA extracted from 100 #1 serum aliquots using primers for the highly conserved 5'NCR of HCV-RNA and 35 amplification cycles. PCR products were analyzed by agarose gel electrophoresis and Southern blot hybridization against a P32-oligonucleotide probe. Sensitivity was evaluated in separate experiments on tenfold dilutions of a reference Chimp serum containing 106 CIDS°/ml. The overall sensitivity of our assay ranged between 102 a n d 103 genome Eq./ml. We establish a semiquantitative score system to evaluate viremia levels: 2 = HCV-RNA levels > 104 genome Eq./ml; 1 =levels between 103 and 104 g.Eq./ml; 0=levels less than 102 g.Eq./ml. The reproducibility of this scoring system was confirmed testing repeatedly in duplicate end-point dilutions of positive serum samples. A statitiscaUy significant relation was observed between elevated HCV-RNA and ALT values (83.8%, chi-square 159.963 P<0.0001). Response to IFN therapy was significantly better in patients with lower baseline HCV-RNA levels. A time relation was found between flare-ups of serum HCV-RNA levels and ALT elevations higher than 3 x normal values with viremia elevations coincident or occuring about 1 month earlier than ALT elevations. This finding suggests that immuno pathogenesis might be responsible of HCV-induced liver damage as in chronic hepatitis B where identical relations were observed between viremia and ALT serum levels. In conclusion, single-step HCV-RNA RT-PCR can be a specific

*Corresponding author. Fax + 39-11-3290726, + 39-11-595588.

SSDI 0928-0197(93)E0027-D

290

and reproducible semiquantitative assay and provides useful diagnostic informations for therapeutic decision making and monitoring of HCV-infected patients.

Introduction Molecular biology techniques exploiting the natural properties of nucleic acids and enzymes which regulate DNA synthesis and replication have become major research tools and are being introduced in the diagnostic laboratory. The improvement in sensitivity provided by the amplification of viral nucleic acids using polymerase chain reaction (PCR) has led to the detection of minute quantities of viruses in unexpected circumstances (Ferre, 1992). This has raised medical scepticism and contributed to overemphasize the risk of false positive results (Ferre, 1992). The most important consequence of the application of these techniques to diagnostics is to use for viruses the same diagnostic criteria used for bacteria which are undetectable in biologic specimens but can be amplified in culture. Culture amplification allows a precise quantitative analysis of bacteria as the number of growing colonies in the culture corresponds to the number of bacteria in the original material. According to this, it is well known that 'sterile' for biological specimens does not mean the absence of bacteria, but a number of them below a well-defined cut-off value. Also for viruses that are now detected with absolute sensitivity we have to rely on quantitative cutoffs for clinical decisions. In spite of a constant belief that PCR is unsuitable for quantitative analysis, there is recent compelling evidence that PCR can become a quantitative assay (Brillanti et al., 1992; Ferre 1992; Farci et al., 1991; Weiner et al., 1990). We have assessed sensitivity, specificity, reproducibility, quantitative analysis potential and clinical significance of a single step reverse-transcriptase and polymerase-chain-reaction method (RT-PCR) for the detection of hepatitis C virus (HCV) RNA.

Patients and Methods Patients

Serum samples (293) were obtained from 50 patients with chronic liver disease and HCV infection treated with 1-6 million units of recombinant alpha interferon (IFN) for 6-12 months. They took part in randomized clinical trials and were followed-up for a mean of 18 months (range 12-45 months) (Negro et al., 1992; Saracco et al., 1991). All of them were anti-HCV positive by 1st and 2nd generation assays (ORTHO HCV ELISA, Ortho Diagnostic Raritan N.J., USA). We analysed at least 5 serum specimens (range 3-16) from each patient, before, during and after therapy. R T - P C R . Total RNA was obtained from 200 #l serum according to Han et al. (1991). Briefly sera were denatured in 5 M guanidinium thiocyanate, 50 mM TRISHC1 (pH 7.5), 25 mM EDTA, 8% v/v 2-mer-captoethanol lysing solution and RNA was recovered by a two-step phenol-chloroform (1:1) extraction and a further extraction with chloroform alone. Samples were precipitated with isopropanol v/v in presence of 300 mM NaOAc and carrier t-RNA. We used as alternative and quicker RNA extraction method, the Chomczynski's and Sacchi's procedure (Chomczynski, 1987). Sera (100 pl) were added to 500 #1 of denaturing solution containing 4 M

291

guanidinium thiocyanate, 25 mM sodium citrate pH 7, 0,5% sarcosyl, 0,1 M 2-mercaptoethanol. Then 50 #1 of 2 M sodium acetate pH 4, 500/~1 of phenol (DEPCwater equilibrated) and 100/~1 of chloroform were added. After centrifugation the upper phase was precipitated with an equal volume of isopropanol in presence of 10 /~g carrier glycogen. The resulting pellet was washed twice with 70% cold ethanol and resuspended in 20/A of DEPC-treated water. RNA was recovered with comparable efficiency using both extraction methods therefore we used the quicker Chomczynsky's and Sacchi's procedure for routine testing. RT-PCR reactions were performed using primers for the 5'-untranslated region of HCV genome that has been shown to be the most conserved region among different HCV isolates (Buck et al., 1992; Choo et al., 1991; Houghton et al., 1992; Kato et al., 1990; Weiner et al., 1990). After denaturation at 70°C. for 5 min 10/~1 RNA aliquots were primed with 150 pmol of oligonucleotide JHC 51 (5'-CCC AAC ACT ACT CGG CTA-3') for c-DNA synthesis (Han et al., 1991). The 25 #1 reaction mixture was incubated for 1 hr at 37°C. with 1 mM dNTPs, 40 units of ribonuclease inhibitor (PROMEGA Corp. USA) and 300 units of M-MLV reverse transcriptase (BRL Corp. USA). After heat-inactivation at 95°C for 5 min and chilling on ice 10/~1 were used for amplification. The 100 #1 PCR reaction contained: 10 mM TRIS-HC1 pH 8,3, 50 mM KC1, 2 mM MgC1 buffer with 200 #M dNTPs, 50 pmoles of sense primer JHC 93 (5'-TTC GCG GCC GCA CTC CAC CAT GAA TCA CTC CCC-3'), 50 pmoles of antisense primer JHC 52 (5'-AGT CTT GCG GCC GCA CGC CCA AAT C-3') and 2,5 units of Taq polymerase (Perkin-Elmer-Cetus Corp. USA). The DNA thermal cycler was programmed for 5 cycles of 2 min denaturation at 95°C, 2 min annealing at 50°C, and 3 min primer extention at 72°C. The following cycles (30) were performed at 95 °C for 1.5 min, 60°C for 1.5 min and 72°C for 2 min. PCR products (10 #1) were analyzed by electrophoresis in a 2% agarose-gel and visualized by ultraviolet fluorescence after ethidium bromide staining. A second-step analysis was performed by Southern blotting of the gels and probing with p32 end-labeled oligonucleotide Alx 89 (5'-CCA TAG TGG TCT GCG GAA CCG GTG AGT ACA-3').

Statistical analysis We used the analysis of variance for repeated measures to estimate the variability between the survival curve of viremia in patients with different baseline levels.

Results

R T-PCR sensitivity and specificity In order to evaluate the sensitivity of the RT-PCR, we tested several times (at least 4 times) in separate experiments serial 10-fold dilutions (in negative human serum) of a reference Chimp serum containing 10 6 CIDS°/ml equivalent to 3 per l07 genomes/ml (g.Eq./ml) (Bradley, 1933; Weiner et al., 1990) (Fig. 1). After hybridization with the pa2-1abeled probe a positive signal could be detected till the 105 dilution of the positive control sample (10 CID) corresponding to about 102- l0 a genome Eq./ml. Therefore the overall sensitivity of our assay ranges between l 0 2 and 10a genome equivalents per ml. The specificity of the results was controlled in each experiment using 2 distilled water samples and 2 HCV-RNA negative sera as negative controls every 20 test specimens in the extraction step. Five additional distilled water control specimens were added in the PCR reaction. The sensitivity and specificity were confirmed by

292

Fig. 1. Agarose gel electrophoresisof 5 repeats of serial dilutions (A,B,C,D,E)of a Chimp serum containing 10 6 CIDso/ml. Lanes A,B,C,D,E contain 10-4, 10 -3 10 -2, 10 and 1 CIDs0/ml respectively. Authoradiography of the same RT-PCR products shown in the agarose gel hybridizedto Alx 89 probe. the results obtained by our assay in a recent RT-PCR HCV-RNA quality control standardization trial for HCV-RNA conducted by the Eurohep pathobiology group (European Study Group on Viral Hepatitis, 1993). The reference panel included 10 serum samples from blood donors with and without HCV infection to test the assay performance and 2 dilution series of sera to evaluate the assay sensitivity. We are 1 of the 5 (out of 31) participant laboratories which performed the entire panel faultlessly.

Semiquantitative analysis of viraemia A score system was used to evaluate semiquantitatively the levels of viremia: 2 = sample positive by both agarose gel electrophoresis and blot-hybridization (corresponding to HCV-RNA levels higher than 104 genome Eq./ml); 1 = sample positive by blot-hybridization but negative by agarose gel electrophoresis (corresponding to a range of serum HCV-RNA levels between 103 and 104 genome Eq./ml); 0 = sample negative by both agarose gel electrophoresis and blot-hybridization (corresponding to less than 102 genomes Eq/ml. of HCV-RNA). Duplicate experiments with 10-fold dilutions of reference Chimp serum confirmed that 0, 1 and 2 values corresponded to less than 102, 103-104 and <104 genomes equivalents per ml of serum respectively (Table 1). The reproducibility of the scoring system was assessed in multiple assays and testing in dublicate end point dilutions of positive samples as shown in Fig. 2. HCV-RNA was detected till the 2nd dilution in serum samples with score = 2 while it was detected till the 1st dilution in specimens with score 1 as shown in Fig. 2.

293 TABLE I Relations among reference chimp serum dilutions and viremia expressed as CIDs0/ml, g.Eq./ml or RTPCR score Chimp Serum dilutions

CIDs0/ml

g.Eq./ml

RT-PCR score

102 103 104 105 106

I0.000 1.000 100 10 1

105-106 104-105 103-104 102-10 a 10-102

2 2 1 1

RNA=

2

1

14A B C D E15A

0

2 B C D16 A B C D

E

C-

Fig. 2. End point dilutions of amplicons from 2 blood samples of patients no. 14 and no. 15, shown in alphabetical order, represent HCV-RNA serum levels at baseline (score = 2) and at the end of interferon therapy (score= 1). End point dilutions of amplicons from blood sample no. 16 corresponding to the serum of another patient with elevated HCV-RNA serum levels before therapy and undetectable levels after the end of treatment. Sera classified a s = 2 were positive at least to the second 10-fold dilution of the amplicons while those classified as = 1 were positive up to the first 10-fold dilution only.

To assess the relations between the variations of viremia and ALT serum levels in patients undergoing interferon therapy, we tested all the blood samples from each given patient in the same PCR assay. We found a significant relation between the levels of HCV-RNA and ALT (Table 2). In 32 patients we studied the time relationships between serum HCV-RNA flare-ups (increments of viremia of at least 1 degree higher than the median value of 3 consecutive monthly tests) and ALT elevations (increments equal to or higher than 3 times the median value of 3 consecutive monthly tests). In 11 cases we observed a decrease of at least 1 point in the viremia score and a rebound to baseline values within 6 month follow-up. In 8 of them (72.7%) viremia rebounds were associated with ALT flare-ups (chi-square= 16.670,

294 TABLE II Relations between HCV-RNA and ALT levels in 293 sera Viremia

0 l

2 Total

ALT (U/l)

Total

< 30

30-45

> 45

number(%)

number(%)

number(%)

58(80.5%) 30(44.8 %) 11(7.2%) 99

9(12.5%) 19(28.3 % ) 14(9%) 42

5(7%) 18(26.9%) 129(83.8%) 152

72 67 154 293

chi-square 159.963 p<0.0001.

P<0.0001). ALT peak values were coincident in 4 cases (50%) with the peak increments of viremia. In the remaining 50% of cases ALT peak values were detected 1 month later than HCV-RNA peak values. In Fig. 3 we represent the survival curves of serum HCV-RNA in interferon treated patients according to baseline viremia levels. Fifty RT-PCR HCV-RNA positive sera were tested also with the bDNA quantitative assay (Branch DNA Assay, Urdea et al., 1991) and a significant correlation (Spearman's correlation, P < 0.006) was found between our semiquantitative score system and bDNA assay values for viremia levels higher than the bDNA assay sensitivity limit (350.000 g.Eq./ml).

4. Discussion

Antibody tests for hepatitis C virus infection are important as screening assays but their detection in the serum does not help to distinguish patients with virus-

80

eo

~

p < 0001

40 ,

~

1

baseline

endoftherlpy

i

J i

6months 12months fo#ow-up

Fig. 3. Survival curve of viremia in 50 interferon-treated patients. Viremia was measured in each patient at baseline, end of therapy, 6 and 12 months of post-treatment follow-up. The results are represented as percent of patients who maintained detectable viremia during the follow-up. Line I corresponds to patients with low baseline viremia (score 1) and Line 2 to those with high baseline viremia (score 2).

295 related disease from healthy carriers of HCV and individuals with past infection (Ulrich et al., 1990). Detection of hepatitis C viral RNA in serum specimens can be helpful for diagnostic and therapeutic purposes (Alberti, 1991; Bonino et al., 1991; Nakatsuji et al., 1992; Oliveri et al., 1990; Ulrich et al., 1990). However, the use of nucleic acid amplification methods which detect viruses with high sensitivity (1-100 molecules of genome equivalents per ml of body fluids) has emphasized the diagnostic concept that detection of the virus does not imply per s e a virus-related disease. In fact minute amounts of virus nucleic acids can be detected in healthy individuals (Gretch et al., 1993; Weiner et al., 1990), therefore quantitation of viremia becomes mandatory to understand the pathogenetic implications of virus replication. Expensive and time-consuming quantitative assays have been proposed using cDNA limiting dilution analysis or competitive PCR (Brillanti et al., 1991; Kaneko et al., 1992). However, two rounds of amplification in the former method makes the quantitation of the initial HCV-RNA more difficult, as shown in the Eurohep RTPCR reliability study (European Study Group on Viral Heapatitis, 1993). On the other hand the addition of a competitive mutant template introduces a new reaction variable (the competition with the target wild-type template) that depends on the templates concentrations. Semiquantitative single-step RT-PCR using the 5'-noncoding region of HC¥-RNA appears a feasible and reproducible technique for measuring HCV-RNA in routine serum specimens. A significant correlation (Spearman's P<0.006) was observed for HCV-RNA levels higher than 350.000 g.Eq./ml between RT-PCR and the bDNA assay. Therefore RT-PCR is mandatory in patients with viremia levels below the bDNA assay detection limit. Single step RT-PCR is safer than nested RT-PCR as for both specificity and contamination problems (Bonino et al., 1991; Gretch et al., 1993; Nakatsuji et al., 1992). The semiquantitative score system appears a practical tool to identify clinically significant variations of viremia. In fact in followed-up patients serum HCV-RNA decrements or rebounds of at least one degree in our score system were associated with remissions and exacerbations of the biochemical activity of liver disease respectively. The significant correlation between serum HCV-RNA and aminotransferase levels could be taken as evidence of direct HCV pathogenicity. However, the type of time relations observed between the fluctuations of viremia and ALT bring into question this hypothesis. In fact serum HCV-RNA rebounds to baseline levels after significant decrements associated with interferon therapy were associated with ALT flare-ups and in 50% of cases ALT peak values were observed in the blood obtained 1 month later than HCV-RNA peak values. These findings are consistent with the time relations reported between HBV-DNA and ALT serum level fluctuations of patients with chronic hepatitis B virus (HBV) infection (Colloredo et al., 1992). Therefore the equilibrium between virus replication and host's antiviral immunoreaction might play a critical role in HCV pathogenesis as well as in that of HBV. The most important clinical application of HCV-RNA RT-PCR appears the monitoring of patients undergoing antiviral therapy (Bonino et al., 1991; Chayama et al., 1991; Colloredo et al., 1992; Gretch et al., 1993; Kanai et al., 1990; Kaneko et al., 1992; Lau and Davis, 1993; Shindo et al., 1991, 1992; Simmonds et al., 1990). The detection of viral nucleic acids becomes very important when the efficacy of antiviral therapy cannot be monitored by immunological markers of virus-induced liver disease, as in the case of HCV-infected patients. The score system can provide a valuable tool to study the relations between baseline viremia levels and response to therapy. Even in our small series of interferon-treated patients, viremia decreased more consistently

296

in patients with low HCV-RNA baseline scores as compared with those with higher scores (Fig. 3). In conclusion single-step RT-PCR using 5'-non coding region primers and a simple semiquantitative score system appears a sensitive, suitable and clinically useful tool for therapeutic decision making and monitoring of HCV-infected patients. Serum HCV-RNA levels appear to decrease significantly in IFN-treated patients who normalize ALT serum levels while viremia rebounds appear to be associated with hepatitis relapses. The persistent clearance of viremia appears more frequent in patients with low baseline HCV-RNA levels.

Acknowledgements Part of this study has been conducted with the support of grants from C.N.R., no. 92.02300.PF39 and I.S.S. no. 26011 between our Laboratory and the Italian National Research Council and Istituto Superiore della Sanita. We wish to thank Dr. Amy Weiner for fruitful discussion of the manuscript and Miss Luciana Biasotto for technical assistance in preparing it.

References Alberti, A. (1991) Diagnosis of hepatitis C: facts and perspectives. J. Hepatol. 12, 279-282. Bonino, F., Brunetto, M.R. and Baldi, M. (1991) Hepatitis C Serology. Eur. J. Gastroenterol. and Hepatol. 3, 580-584. Bradley, D.K. (1993) Human hepatitis viruses. Seminars in Virol. 4, 269-272. Brillanti, S., Garson, J.A., Tuke, P.W., Ring, C., Bruggs, M., Masci, C., Miglioli, M., Barbara, L. and Tedder, R. (1991) Effect of alpha-interferon therapy on Hepatitis C viraemia in community acquired chronic non-A, non-B hepatitis; a quantitative polymerase chain reaction study. J. Med. Virol. 34, 136-141. Buck, J., Purcell, R.H. and Miller, H.R. (1992) Sequence analysis of the 5' non coding region of hepatitis C virus. Proc. Natl. Acad. Sci. 89, 4942-4946. Chayama, K., Saitoh, S., Arase, Y., Ikeda, K., Matsumoto, T., Skai, Y., Kobayashi, M., Unakami, M., Morinaga, T. and Kumada, H. (1991) Effect of interferon administration on serum hepatitis C virus RNA in patients with chronic hepatitis C. Hepatology 13, 1040-1043. Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Analytical Biochemistry 162, 156-159. Choo, Q.-L., Richman, K.H., Han, J.H., Berger, K., Lee, C., Dong, C., Gallegos, C., Coit, D., MedinaSelby, A., Barr, P.J., Weiner, A.J., Bradley, D.W., Kuo, G. and Houghton, M. (1991) Genetic organization and diversity of the hepatitis C virus Proc. Natl. Acad. Sci. 88, 2451-2455. Colloredo, Mels, G., Bellati, G., Leandro, G., Brunetto, M.R., Scorpiniti, A., Fornaciari, G., Borzio, M., Antinozzi, R., Angeli, G., Bonino, F. and Ideo, G. (1992) Reactivation of chronic hepatitis B virus infection: patterns of the fluctuations of HBV-DNA, ALT and IgM anti-HBc (IMx) during the flareups. Hepatology 16, 199A. European Study Group on Viral Hepatitis, Pathobiology Group. (1993) Reliability of polymerase chain reaction for detection of hepatitis C virus. Lancet 341,722-724. Farci, P., Alter, H.J., Wong, D., Miller, R.H., Shin, J.W., Jett, B. and Purcell, R.H. (1991) A long-term study of hepatitis C virus replication in non-A non-B hepatitis. N. Engl. J. Med. 335, 98-104. Ferre, F. (1992) Quantitative or semiquantitative PCR: reality versus myth. PCR Methods and Applications 2, 1-9. Gretch, D.R. et al. (1993) Detection of hepatitis C virus RNA; comparison of one-stage Polymerase Chain Reaction (PCR) with nested-set PCR. Journal of Clinical Microbiology, 31(2), 289-291. Hart, J.H., Shyamala, V., Richman, K.H., Brauer, M.J., Irvine, B., Urdea, M.S., Tekamp-Olson, P., Kuo, G., Choo, Q.-L. and Houghton, M. (1991) Characterization of the terminal regions of hepatitis C viral

297 RNA: identification of conserved sequences in the 5'-untranslated region and poly(A) tails at th 3' end. Proc. Natl. Acad. Sci. USA 88, 1711-1715. Houghton, M., Weiner, A., Han, J., Kuo, G. and Choo, Q.-L. (1991) Molecular biology of the hepatitis C viruses; implications for diagnosis, development and control of viral disease. Hepatology 14, 381-388. Kanai, K., lwata, K., Nakao, K., Kako, M. and Okamoto, H. (1990) Suppression of hepatitis C virus RNA by interferon alpha. Lancet 336, 245-246. Kaneko, S., Murakami, S., Unoura, M. and Kobayashi, K. (1992) Quantitation of HCV-RNA by competitive polymerase chain reaction. J. Med. Virol. 37, 278-282. Kato, N., Yokosuka, O., Omata, M., Hosoda, K. and Ohto, M. (1990) Detection of hepatitis C virus nucleic acid in the serum by amplification with polymerase chain reaction. J. Clin. Invest. 86, 1764-1767. Lau, J.Y.N. and Davis, G.L. (1993) Managing chronic hepatitis C virus infection. Br. Med. J. 306, 469-470. Nakatsuji, Y., Matsumoto, A., Tanaka, E., Ogata, H. and Kiyosawa, K. (1992) Detection of chronic hepatitis C virus injection by four diagnostic systems: first-generation and second-generation enzymelinked immunosorbent assay, second-generation recombinant immunoblot assay and nested polymerase chain reaction analysis. Hepatology 16, 300-305. Negro, F., Baldi, M., Mondardini, A., Ballar6 M., Capalbo, M., Chiaberge, E., Abate, M.L. and Bonino, F. (1992) Continuous vs. intermittent administration of alpha-interferon in chronic hepatitis C. J. Hepatol. 16, $48. Oliveri, F., Baldi, M., Brunetto, M.R., Saracco, G., Rosina, F. Torrani, Cerenzia, M., Rizzetto, M., Soranzo, L., Colla, L., Vallauri, P., Verme, G., Kuo, G., Houghton, M. and Bonino, F. (1990) Antibody to hepatitis C virus in the serum of patients with chronic hepatitis. Eur. J. Gasteronterol. and Hepatol. 2, 347-350. Saracco, G., Baldi, M., Calvo, P.L., Manzini, P., Abate, M.L., Chiaberge, E., Brunetto, M.R., Rizzetto, M., Polito, G., Kuo, G., Chien, D., Houghton, M. and Bonino, F. (1991) Monitoring antiviral non-A, non-B chronic hepatitis therapy by HCV serum markers. J. Hepatol. 13, 5170. Shindo, M., Di, Bisceglie, A.M., Cheung, L., Shill W.-K., Cristiano, K., Feinstone, S. and Hoofnagle, J.H. (1991) Decrease in serum hepatitis C viral RNA during alpha interferon therapy for chronic hepatitis C. Ann. Intern. Med. 115, 700-704. Shindo, M., Di Bisceglie, A.M. and Hoofnagle, J.H. (1992) Long-term follow-up of patients with chronic hepatitis C treated with alpha-interferon. Hepatology 15, 1013-1016. Simmonds, P., Zhang, L.Q., Watson, H.G., Rebus, S., Ferguson, E.D.., Balfe, P., Leadbetter, G.H., Yap, P.L., Peutheber, J.F. and Ludlam, C.A. (1990) Hepatitis C quantification and sequencing in blood products, haemophiliacs and drug users. Lancet 336, 1469-1472. Ulrich, P.P., Romeo, J.M., Lane, P.K., Kelly, I., Daniel, L.J. and Vyas, G.N. (1990) Detection, semiquantitation and genetic variation in hepatitis C virus sequence amplified from the plasma of blood donors with elevated alanine aminotransferase. J. Clin. Invest. 86, 1609-1616. Urdea, M.S., Horn, T., Fultz, T.J., Anferson, M., Running, J.A., Hamren, S., Ahle, D. and Chang, C. (1991) Branch DNA amplification multimers for the sensitive, direct detection of human hepatitis viruses. NAR 24, 197-200. Weiner, A.J., Kuo, G., Bradley, D.W., Bonino, F., Saracco, G., Lee, C., Rosenblatt, J., Choo, Q.-L. and Houghton, M. (1990) Detection of hepatitis C viral sequences in non-A, non-B hepatitis. Lancet 335, 1-3.