Third-generation hepatitis C virus tests in asymptomatic anti-HCV-positive blood donors

Copyright © Journalof Helmtolog.r 1994

Journal of Hepatology 1994; 21:730-734 Prhlted in Denmark. All rights reserved Munksgaard. Copenhagen

Journal of Hepatology ISSN 0168-8278

Third-generation hepatitis C virus tests in asymptomatic anti-HCV-positive blood donors A n t o n i o Craxi l, M a r i o Valenza 2, C a r m e l o F a b i a n o 2, Silvio M a g r i n 2, G e r m a n a Fiorentino 2, Orazia D i q u a t t r o I and Luigi Pagliaro ~ t lstituto Medicina Generale, UniversiO' of Pah, rmo. and 2Divisione di Medicina h~terna. Ospedah, V. Cervello. Pah, rmo. Italy

(Received 26 June 1993)

This study evaluated the performance of third-generation anti-HCV assays in blood donors who were positive by secondgeneration anti-HCV, and assessed any possible relationship between antibody patterns, HCV replication and liver damage. Fifty-two second-generation enzyme immunoassay-positive asymptomatic Italian blood donors were retested for anti-HCV by third-generation enzyme immunoassay and recombinant immunoblot assay (Ortho third-generation enzyme immunoassay, third-generation recombinant immunoblot assay), utilising recombinant C33c and NS5 and synthetic peptide C100 and C22 antigens, and for HCV-RNA by "nested" polymerase chain reaction with 5' region primers. Alanine aminotransferases were tested monthly for 6 months. Two out of 52 second-generation enzyme immunoassay-positive donors were third-generation enzyme immunoassay, third-generation recombinant immunoblot assay and HCV-RNA negative. Among 50 third-generation enzyme immunoassay-positive cases, two had a third-generation enzyme immunoassay optical density-l: six were third-generation recombinant immunoblot assay negative (one HCV-RNA+ve), eight "indeterminate" (two HCV-RNA+ve) and 34 positive (22 HCV-RNA+ve). All "indeterminate" subjects reacted only to C22. HCV-RNA was positive in 22/34 cases with positive third-generation recombinant immunoblot assay (two or more Ags), 3/9 "indeterminate" and 1/11 negative. Alanine amino-transferases were abnormal in 13 cases with positive third-generation recombinant immunoblot assay, one was "indeterminate" and three were negative. There was, however, a significant relation between C33c positivity and raised alanine aminotransferases (p<0.023). Reactivity to C33c was strongly related to viraemia; 21/29 C33c-positive cases were also HCV-RNA positive, as opposed to 5/25 C33c negative. Viraemia was often but not always associated with liver damage, alanine aminotransfoase levels being abnormal in 13/26 (50%) HCV-RNA positive and 4/28 (14%) HCV-RNA negative subjects (p<0.005). Third-generation HCV tests can reduce the number of second-generation enzyme immunoassay positive donors with a negative HCVRNA. The concordance between third-generation enzyme immunoassay and third-generation recombinant immunoblot assay is high (96°/,,). Some third-generation enzyme immunoassay/third-generation recombinant immunoblot assay positive, HCV-RNA-negative subjects might have latent viraemia. An "indeterminate" third-generation recombinant immunoblot assay pattern or even a negative third-generation recombinant immunoblot assay does not exclude active infection, since HCV-RNA is often present. © Journal of Hepatology. Key words: Blood donors; Chronic hepatitis; EIA; HCV-RNA; Hepatitis C virus; Liver disease; RIBA; Third-generation

tests

Testing by multiple-antigen assays has increased the rate of detection of hepatitis C virus (HCV) infection among blood donors (1) and patients with chronic liver disease

of suspected non-A, non-B aetiology (2,3). This increased sensitivity will eventually allow a reduction in the rate of post-transfusion HCV infection in the USA to about 1 in

Correspondence to: Prof. Antonio Craxi, Div. di Medicina, Ospedale V. Cervello, Via Trabucco 180, 90100 Palermo, Italy.

THIRD-GENERATION HCV TESTS IN BLOOD DONORS every 6000 blood units (4), but will still not entirely eliminate the risk of transfusion-related hepatitis C (5). Some HCV-infected individuals might in fact have HCV viraemia (as detected by HCV-RNA testing by polymerase chain reaction (PCR)) in the absence of antibodies detectable by second-generation tests (6,7), because they are either in a very early phase of infection or are not expressing an antibody response against some epitopes. Antibody-negativeHCV infection is uncommon among subjects with chronic liver disease (3,8), in whom HCV replication (HCV-RNA in serum) is almost always confirmed by the positivity of the second-generation enzyme-linked immunosorbent assay (EIA2) and of at least one antibody in the recombinant immunoblot (RIBA2) assay. A major problem for blood banks and in the aetiologic diagnosis of liver disease is the "indeterminate" (i.e. reactive only for one HCV antibody) RIBA2 assay. Although this is said to indicate mostly a false-positive reaction (9), 21-24% of RIBA2 "indeterminate" blood donors are actually HCV-RNA positive (10,11). Most "indeterminate" donors have antibodies to a viral nucleocapsid (C22) epitope (12). Conversely, a positive RIBA2 test in a blood donor (i.e. reactive for two or more antibodies) strongly suggests active infection, since 76 -95% of these subjects are HCV-RNA seropositive by PCR (6,10-12). New HCV tests are being developed to reduce "indeterminate" results and to obtain a test which will detect current infection without the need for further confirmatory tests. While some future assays will incorporate chymeric polyproteins with enhanced epitope display (13) or synthetic peptides from core and non-structural proteins (14), third-generation assays (EIA3 and RIBA3, Ortho Diagnostics, Raritan, New Jersey, USA) evolving from the currently available assay are undergoing clinical assessment (9,15,16). The RIBA3 test utilises two recombinant HCV antigens (C33c produced in E. coli, NS-5 produced in yeast) and two synthetic peptide bands, derived from the nucleocapsid (C22) and the non-structural (C100) regions of the HCV genome. Three recombinant HCV-encoded antigens are instead used in the EIA3 (C22; C200, a chymeric protein encompassing C33c and C100 epitopes; NS-5). Initial evaluations (15,16) suggest a high rate of resolution of "indeterminate" samples and a marginally higher detection rate of HCV infection. GarciaSamaniego and coworkers (18) have shown an excellent test performance in 20 RIBA2-indeterminate high-risk subjects, most of whom had abnormal liver tests. To evaluate the performance of the third-generation Ortho anti-HCV assays in blood donors who are positive by second-generation anti-HCV, we evaluated a cohort of subjects referred to our Institution from local Blood Banks. We also assessed the possible relationship between

731 RIBA3 antibody patterns, HCV replication as evaluated by HCV-RNA and biochemical evidence of liver damage.

Subjects and Methods Patients The study group consisted of 52 anti-HCV positive, anti-HIV-1/2 negative asymptomatic Southern Italian blood donors. These cases were selected from a cohort of 256 subjects, who were consecutively referred for evaluation to our Liver Clinic between 1 January 1990 and 31 December 1992, solely on the basis of the availability of serum samples and of at least 6 months of follow up. All these volunteer blood donors had been found anti-HCV positive at one of three local Blood Banks by first- or second-generation EIA, and had been excluded from making further donations. The male/female ratio was 39/ 13, and mean age 42 years ( S D ± l l . 4 , range 19-64). All denied intravenous drug-addiction or male homosexuality. Forty-two subjects (80.8%) were first-time donors. Two cases (3.8%) had a history of previous acute hepatitis, and five others (9.6%) had an isolated previous finding of "raised aminotransferases". Forty,six (88.6%) had had minor dental care in the past, and 27 (51.9%) previous surgery, in seven cases with blood transfusion. A family history of liver disease (i.e. at least one first-degree relative with chronic hepatitis or cirrhosis) was found in 13 subjects (25%). Methods Study patients were retested by second-generation EIA and RIBA (EIA2/RIBA2; Ortho Diagnostics, Raritan, New Jersey, USA). Their anti-HCV status was reassessed by third-generation EIA and RIBA (EIA3, RIBA3; Ortho Diagnostics, Raritan, New Jersey, USA), utilising recombinant C33c and NS5 and synthetic peptide C100 and C22 antigens. HCV-RNA was evaluated in serum, on the same samples tested for anti-HCV, by "nested" PCR with 5'UTR primers, as reported elsewhere (17). All tests were performed in duplicate, in a single batch. The detection limit of our PCR was estimated by comparison to a reference standard (kindly provided by Dr. Jang Han, Chiron, Co. Emeryville, California, USA) to be about 0.5x 103/ml HCV-RNA genome/equivalents. Alanine aminotransferases (ALT) were tested bimonthly for at least 6 months; elevations above the upper limits of normal range were classified as "abnormal" only if present on at least two separate checks. Statistics Data were analysed by the two-sided Fisher's exact test or by chi-square (corrected or overall), as appropriate.

732

A. CRAX| et al.

Results All study subjects were EIA2 positive; R I B A 2 was positive (i.e. reactive with two or three antigens) in 26 cases, "indeterminate" (i.e. reactive for only one antigen) in 14 and negative in 10. Two out o f 52 (3.8%) EIA2-positive d o n o r s were antiH C V negative by EIA3 (Table l). They were RIBA3 and H C V - R N A negative. Both had low optical density ( O D < I ) by EIA2, reacted weakly (l + ) to C22 by RIBA2, and had n o r m a l ALT values. A m o n g the 50 EIA3-positive cases (Table 1), two had an E I A 3 0 D - < I : one o f them was RIBA3 and H C V - R N A negative, and the other was RIBA3 "indeterminate" (C22 only) and H C V - R N A positive. The remaining 48 cases had an E I A 3 0 D > l: six were RIBA3 negative (one H C V - R N A positive), eight "indeterminate" (two H C V - R N A positive) and 34 positive (22 H C V - R N A positive). All "indeterminate" samples reacted only for C22 (l + to 3+). Overall, all RIBA2 positive cases were confirmed as positive by RIBA3. Five out o f 14 R I B A 2 "indeterminate" subjects were resolved by RIBA3 (two negatives; three positives, one o f whom was HCVR N A positive). The remaining nine "indeterminate" R I B A 2 subjects remained the same on RIBA3 testing, three being H C V - R N A positive. Main features o f the study subjects, according to their RIBA3 status, are shown in Table 2. Reactivities by RIBA3 for each pattern o f antibodies to H C V are shown in Table 3, related to H C V - R N A and alanine aminotransferase status. Overall, H C V - R N A was

present in serum in 22/34 cases (65%) with a positive (two or more Ag) RIBA3, in 3/9 (33%) "indeterminate", and in 1/9 (11%) RIBA3-negative cases (p<0.001). A L T levels were a b n o r m a l in 13 subjects (38%) with positive RIBA3, one (11%) "indeterminate" and two (22%) negative (p not significant). C100 reactivity was found in 30 subjects (19 o f whom were H C V - R N A positive), C33c in 29 subjects (21 HCVR N A positive), C22 in 43 subjects (25 H C V - R N A positive) and NS5 in six subjects (five H C V - R N A positive). Reactivity to C33c was strongly related to viraemia, 21/ 29 C33c-positive cases being also H C V - R N A positive as opposed to 5/25 C33c negatives (p=0.000). Correlations between other reactivities and R N A were weaker (C22: p<0.004; C100: p<0.013; NS5: p<0.08). H C V viraemia was not always associated with biochemical evidence o f liver damage, A L T being n o r m a l in 13/26 (50%) HCV-RNA-positive patients. F o u r of the 26 HCVRNA-negative subjects also had a b n o r m a l ALT. One o f them, who was EIA3 and RIBA3 negative, had active HBV infection.

Discussion Testing by EIA3 alone eliminated some presumably false-positive E I A 2 / R I B A 2 results, as already reported by Garcia-Samaniego et al. (18). Two subjects (4% o f the entire population) classified as p r o b a b l y H C V infected by second-generation tests were shown to be non-reactive by EIA3 and RIBA3, and were also H C V - R N A seronegative.

TABLE 1 Correlation between anti-HCV by EIA3 and RIBA3 in 52 EIA2/ RIBA2 anti-HCV positive blood donors RIBA3

TABLE 3 Hepatitis C virus antibodies patterns by RIBA3 according to HCVRNA and alanine aminotransferase level

Negative

"Indeterminate"

Positive

C22

C33c

CI00

NS5

HCV-RNA All cases (pts>ALT)**

2 1 6

0 1 8

0 0 34

+ +

+ +

+ +

+ +

+ -

5 (2) 1 (I)

OD=optical density.

+ +

+ +

+ +

-

+ -

13 (7) 6(1)

TABLE 2

+ +

+ +

-

-

+ -

3 (2) 1 (0)

EIA3

Negative OD- 1

Main features of subjects in respect to RIBA3 status

Mean age (yrs) M/F Raised ALT HCV-RNA +ve Acute hepatitis Major surgery Dental care Transfusions

+

-

+

-

+

I (0)

RIBA3 +ve (34 pts)

RIBA3 indet. (9 pts)

RIBA3 -ve (9 pts)

+

-

+

-

-

4 (0)

+

-

-

+

3 (1)

46 24/I 0 14/34 22 1 21 28 4

45 7/2 1/9 3 0 2 8 0

25 8/1 I/9 1 1 3 8 1

+

.

. .

. .

-

. .

.

.

. .

6 (0) +

.

.

1 (1) 8 (I)* Total 52 (16)

* HBsAg-positive, HBeAg-negative, HBV-DNA-positive. ** Patients with alanine aminotransferase levels above upper normal values.

THIRD-GENERATION HCV TESTS IN BLOOD DONORS More importantly, a significant p r o p o r t i o n (5 out o f 14) o f the R I B A 2 "indeterminate" d o n o r s were reassigned by RIBA3 to true H C V infection or false positives (none viraemic). However, most RIBA2 "indeterminate" subjects had similar test results by RIBA3. Since one out o f three o f them is H C V - R N A positive, it does not seem reasonable to classify these subjects as possible false positives as suggested by the manufacturer. Thus, in accordance with the results obtained by R I B A 2 (10-12), the "indeterminate" class should be avoided in reporting test results. In a blood d o n o r population a strong reactivity by EIA3 is by no means indicative of a positive RIBA3 result. Thus c o n f i r m a t o r y testing should still be performed in these cases. Due to the intrinsically reduced sensitivity of R I B A c o m p a r e d to EIA, a small p r o p o r t i o n (0.5% in our study) of RIBA3 negative, H C V - R N A positive subjects may still be found. Hence an EIA3-positive blood donor, whatever his EIA3 optical density or RIBA3 status, should not donate blood since he/she is potentially infectious (19). The a n t i b o d y reactivities most strongly related to HCVR N A positivity were C22 and C33c. Although C22 was present in a larger number of H C V - R N A positive cases, it was quite often also found in non-viraemic patients. Thus its association with viral replication was statistically less significant than that o f C33c. These data, which are similar to the results o f Martinot-Peignoux et al. (20), differ from Waumans et al. (12), who reported that C22 is the best correlate to active infection. However, this same group found, in a different study (21), that many C-22 positive patients were non-viraemic. NS-5 reactivity, associated in most cases with H C V - R N A seropositivity, is too u n c o m m o n to be useful. In fact, none of the RIBA3 patterns is predictive o f H C V - R N A status, although the simultaneous presence o f C33c and/or C22 (22, 23) and/ or NS-5 is strongly suggestive o f viraemia. Biochemical evidence o f liver damage was strongly related, in this population o f apparently healthy individuals, to the presence o f H C V - R N A , as already reported by Sank a r y et al. (7). A few donors, however, had HCV viraemia in the absence of raised ALT. This again raises the question o f a possible state of healthy long-term viral carriage (24,25). However, even when testing for HCV viraemia using a sensitive "nested" PCR, some anti-HCV-positive, HCV-RNA-negative subjects might have subliminal viraemia. "Occult" or discontinuous HCV replication is probably the cause of a b n o r m a l ALT in the three out of 26 HCV-RNA-negative individuals who do not have other causes of hepatocellular necrosis. Repeated testing for H C V - R N A at different points in time might identify fluctuations in HCV replication, thus reducing the number of false-negative cases.

733

Third-generation HCV tests can reduce the number o f E I A 2 / R I B A 2 positive donors with negative HCV-RNA. The concordance between EIA3 and RIBA3 is high (96%), but a large p r o p o r t i o n of EIA3/RIBA3-positive cases are still HCV-RNA-negative. Some may represent previous, healed infection, but others (especially those with raised ALT and/or C33c positivity) might have latent viraemia. The "indeterminate" pattern should be considered as a true positive, since H C V - R N A is often present. A negative RIBA3 is no absolute p r o o f o f false positivity o f EIA3, as H C V - R N A can also be found in these cases.

Acknowledgements This work was partly funded by grant n.91.04202.ST75 Consiglio Nazionale delle Ricerche. Dr. Tommaso Bottalico (Ortho Diagnostics, Italy) kindly provided EIA3 and RIBA3.

References I. Van der Poel HL, Cuypers HTM, Reesink HW, et al. Confirmation of hepatitis C virus infection by new four-antigen recombinant immunoblot assay. Lancet 1991; 337: 317-9. 2. Craxi A, Fiorentino G, Di Marco V, et al. Second-generation tests in diagnosis of chronic hepatitis C. Lancet 1991; 337: 1354. 3. Chemello L, Cavalletto 19, Pontisso E et al. Patterns of antibodies to hepatitis C virus in patients with chronic non-A, non-B hepatitis and their relationship to viral replication and liver disease. Hepatology 1993; 17: 179-82. 4. Kleinman S, Busch M, Holland P. Post-transfusion hepatitis C virus infection. N Engl J Med 1992; 327: 1601. 5. Nelson KE, Donahue JG, Stambolis V. Post-transfusion hepatitis C virus infection. N Engl J Med 1992; 327: 1601-2. 6. Romeo JM, Ulrich PE Busch ME Vyas GN. Analysis of hepatitis C virus RNA prevalence and surrogate markers of infection among seropositive voluntary blood donors. Hepatology 1993; 17: 188-95. 7. Sankary T, Romeo J, Ulrich E Busch M, Vyas GH. Detection of HCV in seronegative donors with raised ALT. Lancet 1992; 340: 249-50. 8. Nakatsuji Y, Matsumoto A, Tanaka E, Ogata H, Kiyosawa K. Detection of chronic hepatitis C virus infection by four diagnostic systems: first-generation and second-generation enzymelinked immunosorbent assay, second-generation immunoblot assay and nested polymerase chain reaction analysis. Hepatology 1992; 16: 300-5. 9. Irving WL, Day S, Eglin RE et al. Hepatitis C in blood donors. Lancet 1993; 341: 835-6. 10. Yun Z-B, Lindh G, Weiland O, Johansson B, Sonnerborg A. Detection of hepatitis C virus (HCV) RNA by PCR related to HCV antibodies in serum and liver histology in Swedish blood donors. J Med Virol 1993; 39: 57~1. 11. Follen EAC, Dow BC, McOmish E et al. HCV confirmatory testing of blood donors. Lancet 1991; 338: 1024. 12. Waumans L, Claeys H, Verhaert H, Mertens W, Vermylen C. Hepatitis C virus confirmation in blood donor screening. Vox Sang 1993; 64: 145-9. 13. Chien DY, Choo Q-L, Tabrizi A, et al. Diagnosis of hepatitis C virus (HCV) infection using an immunodominant chymeric polyprotein to capture circulating antibodies: reevaluation of the

734

14.

15. 16.

17.

18.

19.

A. CRAXi et al. role of HCV in liver disease. Proc Natl Acad Sci USA 1992: 89: 1001 I-5. Ishida C, Matsumoto K, Fukada K, Matsushita K, Shiraki H, Maeda Y. Detection of antibodies to hepatitis C virus (HCV) structural proteins in anti-HCV positive sera by an enzymelinked immunosorbent assay using synthetic peptides as antigens. J Clin Microbiol 1993: 31: 936-40. Proceedings of the Fourth International Symposium on HCV, Tokyo May 7-9, 1993. Proceedings of the International Symposium on Viral Hepatitis and Liver Disease (The 8th Triennial Congress). Tokyo May 1014, 1993. Magrin S, Craxi A, Fabiano C, et al. Serum HCV-RNA and response to alpha-interferon in anti-HCV positive chronic hepatitis. J Med Virol 1992: 38: 200-6. Garcla-Samaniego J, Enrlquez A, Soriano V, Gutierrez M. Baquero M, Munoz E Third-generation recombinant immunoblot assay to confirm hepatitis C virus-indeterminate serological samples. Vox Sang 1993: 64: 191-2. Halfon E Rousseau S, Tamalet C, et al. Indeterminate second-

20.

21.

22. 23.

24.

25.

generation hepatitis C recombinant immunoblot test: detection of hepatitis C virus infection by polymerase chain reaction. J lnectf Dis 1992: 166: 449. Martinot-Peignoux M, Marcellin E Xu L-Z, et al. Reactivity to c33c antigen as a marker of hepatitis C virus multiplication. J Infect Dis 1992: 165: 595-6. Claeys H, Volkaerts A, Verhaert H, De Beenhouwer H, Vermylen C. Evaluation of anti-HCV capsid indeterminate serum samples. Lancet 1992; 340: 249. Irving WL, Day S, Eglin R,P et al. HCV and PCR negativity. Lancet 1992; 339: 1425. Allain JR Rankin A, Kuhns MC, McNamara A. Clinical importance of HCV confirmatory testing in blood donors. Lancet 1992; 339: 1171-2. Alberti A, Morsica G, Chemello L, et al. Hepatitis C viraemia and liver disease in symptom-free individuals with anti-HCV. Lancet 1992: 340: 697-8. Brillanti S, Foli M, Gaiani S. Masci M. Miglioli M, Barbara L. Persistent hepatitis C viraemia without liver disease. Lancet 1993; 341: 464-5.