Mixed cryoglobulinemia and hepatitis C virus

Mixed Cryoglobulinemia

and Hepatitis C Virus

PATRICECACOUB, M.D., FRANCOISELUNEL FABIANI, M.D., Ph.D., LUCILE MUSSET, Ph.D., MICHELE PERRIN, LIONNEL FRANGEUL,JEAN MARC LEGER, M.D., JEAN-MARIE HURAUX, M.D., Professor, JEAN-CHARLES PIETTE, M.D., Professor, PIERRE GODEAU, M.D., Professor, Paris. France

Mixed cryoglobulinemia (MC) is frequently associated with clinical and biological evidence of liver disease and has recently been reported in cases of hepatitis C virus (HCV) infection. The aim of this study was to assess prospectively in a large series of MC patients: (1) the prevalence of HCV markers (anti-HCV antibodies and HCV RNA in serum and cryoprecipitate); (2) the main clinical, biologic and liver histologic features in patients with or without HCV infection. PATIENTS: One hundred fifteen consecutive unselected MC patients were studied: 45% had well-defined underlying diseases (“nonessential” MC). Fifty-five percent with no cause of MC were considered to have “essential” MC and were subjected to indepth examination. METHODS: Patients were considered to have MC if two successive determinations of their serum cryoglobulin level were above 0.05 g/L. Anti-HCV antibodies (Ab) were detected in all patients by second-generation tests (ELISA, RIBA). We also looked for HCV RNA sequences amplified by polymerase chain reaction (PCR) in the sera and cryoprecipitates of 39 patients; HBs antigen, anti-BBs Ab and anti-HBc Ab in all patients; and HBV DNA in 20 sera and 17 cryoprecipitates. Quantitative HCV Ab and RNA studies were performed on whole serum, cryoprecipitates, and supernatants. Clinical features were recorded retrospectively for BACKGROUND:

From the Department of Internal Medicine (PC, J-CP, PG), Vrrology Laboratory (FLF, MP, LF, J-MH), Immunochemistry Laboratory (LM), Department of Neurology (JML), Hopital La Pitie-Salpbtriere, 83, Boulevard de I’Hopital, 75013 Paris, France. Presented in part at the 56th Annual Scientific Meeting of the American College of Rheumatology, Atlanta, Georgia, October 1992. Reprint requests should be addressed to Patrice Cacoub, M.D., Department of Internal Medicine, Hopital La Pitie-Salpetriere, 83, Boulevard de I’Hopital, 75013 Paris, France. Manuscript submitted October 14, 1992, and accepted in revised form June 2, 1993.

124

February

1994

The American

Journal

of Medicine

Volume

each HCV tive with

patient. Liver biopsies from 23 antiAb-positive and 7 anti-HCV Ab-negapatients were examined histologically, qualitative and quantitative analysis. RESULTS: Anti-HCV Ab were found in 47/ 115 (41%) patients by ELISA and RIBA: 33/63 (52%) essential MC and 14/52 (27%) nonessential MC. Among the 63 essential MC patients, the 33 anti-HCV Ab-positive (Group 1) were compared to the 30 antiHCV Ab-negative patients (Group 2). Group 1 patients had more cutaneous involvement (Raynaud’s phenomenon, purpura, livedo, distal ulcers, or gangrenous changes) (17 versus 5: p = 0.004), higher alanine aminotransferase levels (110 + 22 versus 41 +- 10 IU; p < 0.005), higher serum cryoglobulin levels (0.35 ? 0.07 versus 0.12 +- 0.04 g/L; p = O.Ol), lower CH50 (28 +- 3 versus 44 + 2 CH50/mL; p = 0.0001) and lower C4 levels (0.20 -t 0.02 versus 0.29 * 0.03 g/L; p < 0.04). The prevalence of HBV serum markers was low in both groups, and HBV DNA was never detected in any of the sera and cryoprecipitates tested. HCV RNA sequences were detected in lo/16 (63%) sera and 12/16 (75%) cryoprecipitates from Group 1 patients, whereas they were not in the sera or cryoprecipitates from 23 Group 2 patients. Using quantitative PCR, HCV RNA in cryoprecipitates was concentrated 20 to 100 times despite the absence of significant anti-HCV Ab concentration in these samples. Histologic examination of liver biopsies revealed a spectrum of lesions ranging from chronic active hepatitis to cirrhosis, but Knodell’s score did not differ between the groups. CONCLUSION: (1) About 50% of the essential MC patients had anti-HCV Ab, and these patients had more severe cryoglobulinemiaassociated clinical and biological signs; (2) HCV RNA sequences were found in the large majority of sera and cryoprecipitates from patients with essential MC and anti-HCV Ab and were more concentrated in cryoprecipitates than in supernatants. These results

96

MIXED

CRYOGLOBULINEMIA

suggest a role for HCV in the pathogenesis of MC and indicate that many cases of essential MC may be secondary to HCV infection and thus nonessential.

T

he syndrome of mixed cryoglobulinemia (MC) is characterized by the clinical triad of purpura, arthralgia, and asthenia associated with cryoglobulins composed of different immunoglobulins, with a monoclonal component in type II and only polyclonal immunoglobulins in type III [1,2]. The possibility that the disorder represents the consequence of an immune-complex-type vasculitis is supported by the clinical features, the analysis of the cryoglobulins, the usually depressed level of complement during the active phase of the disease, and the deposition of immunoglobulins and complement in the lesions. Possible mechanisms include specific immune interactions with viral antigens, perhaps cross-reacting with the host, polyclonal activation of B and/or T cells, host genetic makeup and other environmental cofactors. The antigens that trigger the production of antibodies allowing for the subsequent formation of immune complexes are, however usually unknown. MC is frequently associated with clinical and biological evidence of liver disease [2,31. Several studies found evidence of liver dysfunction in 60% to 80% of patients with essential MC whose liver biopsies showed triaditis, chronic active hepatitis, or cirrhosis [3-W. The association between MC and chronic liver disease is also strong when cryoglobulins are sought in patients with chronic liver disease and they have been found in 6% to 43% of such patients 161. There has been some controversy about which comes first, MC or chronic liver disease, but it seems fairly clear that MC is often a manifestation of underlying chronic active or persistent hepatitis. The hepatotropic antigens that can trigger production of MC have been sought for many years. After initial positive studies by Levo et al [71 and others [4,6,81, numerous further studies did not find any relationship between hepatitis B virus (HBV) infection and MC [5,9-121. Since the recent demonstration that hepatitis C virus (HCV) infection is responsible for about 90% of the cases of non-A, non-B hepatitis [131 and can give rise to extrahepatic manifestations [141, some authors have reported cases of HCV infection in patients with MC, usually as single case reports, suggesting a possible role for HCV in MC pathogenesis [14171. Only three studies on large populations [18-

February

AND

HEPATITIS

C VIRUS

/ CACOUB

ET AL

201 found anti-HCV antibodies in about half of patients with MC, but in these studies, patients were highly selected, all of Italian origin (the prevalence of HCV infection is particularly high in Italy), and many of them also had serologic evidence of HBV infection. In a preliminary study, we also found that 50% of the patients with essential MC had serologic evidence of HCV infection [21]. In order to assess the exact role of HCV in the pathogenesis of MC, we conducted a prospective study analyzing in a large series of unselected MC patients: (1) the prevalence of HCV markers (antiHCV antibodies and HCV RNA) in sera and cryoprecipitates; (2) the main clinical, biological, and liver histological features in patients with or without HCV infection markers.

PATIENTS One hundred and fifteen unselected patients with MC were seen consecutively between January 1991 and April 1992 in the Internal Medicine and Neurology Departments of our university hospital. Of these, 52 patients had one or more “classic” causes of MC: autoimmune disorder (n = 33), nonviral infectious disease (n = ll), or malignant hematologic disorder (n = 10). The 63 patients without wellldefined underlying disease were considered to have “essential” MC and were studied in depth.

METHODS Anti-HCV Antibodies and Other Viral Markers All sera were kept frozen at - 80°C until assayed for anti-HCV antibodies (Ab), according to the manufacturers’ instructions, using second-generation enzyme-linked immunosorbent assay (ELISA, Ortho Diagnostic Systems, Roissy, France) and recombinant-base immunoblot assay (RIBA Chiron, Emeryville, CA, Diagnostic Systems). The HCV ELISA detects Ab directed against structural (C22-3) and/or nonstructural (ClOO-3,620O) HCV antigens, and the results are expressed as a ratio (optical density/cutoff). Samples are considered reactive when the ELISA ratio is repeatedly above 1. The RIBA involves four recombinant HCV antion nitrocellulose gens immobilized strips (structural = C22-3, nonstructural = (X-1-1, ClOO-3, C33c) produced in yeast as fusion proteins with superoxide dismutase. Low and high IgG bands are used as positive controls. Superoxide dismutase is used as a negative control to eliminate false-positive results due to the presence of antisuperoxide dismutase Ab in the test sera. The RIBA is considered reactive if two or more bands are

1994

The American

Journal

of Medicine

Volume

96

125

MIXED

CRYOGLOBULINEMIA

AND

HEPATITIS

C VIRUS

/ CACOUB

observed, indeterminate if only one band is present, and nonreactive if there is no band with an intensity greater than that of the low positive control. To determine whether any of the concentration of anti-HCV Ab was higher in the cryoprecipitate than the serum, HCV Ab were quantitatively measured in serial two-fold dilutions of both samples from 6 HCV Ab-positive patients with essential MC. The anti-HBc Ab concentration was also measured in parallel in sera and cryoprecipitates as controls. Sera were also tested for the presence of hepatitis B surface antigen (HBs Ag), anti-HBs Ab and anti-hepatitis B core Ab (HBc Ab) using commercial immunoassays (Abbott Laboratories, Abbott Park, IL; Diagnostic Pasteur). HBV DNA was sought in all sera by means of molecular hybridization (Genostics, Abbott Laboratories). Detection of HCV RNA and HBV DNA Using the Polymerase Chain Reaction (PCR)

HCV PCR was performed on sera and cryoprecipitates that had been kept at -80°C and previously unthawed, to avoid false negativity due to RNA destruction by RNases and false positivity due to contamination. Serum or cryoprecipitate RNA was extracted, reverse transcribed, and amplified as previously described [22]. Serum or cryoprecipitate was diluted to a final volume of 350 FL in 0.05 M tris-hydrochloric acid (HCl) pH 7.5/0.01 M ethylenediamine tetra acetic acid (EDTA)/O.l M sodium chloride/l% sodium dodecyl sulfate and digested with 200 pg/mL of proteinase K. Protein was removed by extraction with phenol/chloroform. RNA was then precipitated by the addition of 2.5 volumes of ice-cold absolute ethanol. The pellet was dissolved in 30 PL of sterile distilled water (RNAsin, Promega, Madison, WI). For cDNA synthesis, a lo-FL mixture was prepared with 8 ~.LLof RNA solution and 10 pmoles of the antisense synthetic oligonucleotide (outer SRI). This mixture was heated at 100°C for 2 minutes then quickly chilled on ice. cDNA synthesis was carried out according to the manufacturer’s instructions in 1 x RT buffer supplemented with 0.01 dithiothreitol M (DTT) (Gibco, BRL), 200 units of cloned Moloney murine leukemia virus reverse transcriptase (Gibco, BRL), 1 mM of each dNTP and 200 units of RNasin in a final volume of 20 FL incubated at 37°C for 45 minutes. For amplification, primers located in the 5’ noncoding highly conserved sequence of HCV were used. PCR was carried out in a 5O-pL mixture containing 1 x PCR

126

February

1994

The American

Journal

of Medicine

Volume

ET

AL

buffer, 125 p,mol/L of each dNTP, 8 pM of each inner primer (SR2-SF2), 2.5 units of recombinant Taq DNA polymerase (Perkin-Elmer-Cetus) and one fourth of the cDNA sample. After an initial 5 minutes denaturation at 94°C the reaction mixture was subjected to 40 cycles at 94°C for 45 seconds, 45°C for 45 seconds and 72°C for 45 seconds, and a last cycle at 94°C for 45 seconds, then 45°C for 45 seconds and 72°C for 5 minutes. One tenth of the amplification mixture was then separated electrophoretically in a 2% agarose gel and stained with ethidium bromide. To assess the specificity of the PCR products, Southern blotting was performed under the usual stringent conditions, using a radiolabeled oligonucleotide probe located between the PCR primers. All experiments were performed in parallel with: (1) serum-free lysis buffer to detect carryover at each step of the procedure (extraction and cDNA step), (2) the reaction mixture without DNA (PCR step), and (3) negative sera. Each sample was tested in at least two different series. A quantitative PCR assay was performed in three cases. Briefly, RNA extracted from sera, supernatants, and cryoprecipitates were reverse transcribed and amplified in parallel by means of nested PCR after serial five-fold dilutions. The signals obtained after electrophoresis in an agarose gel and hybridation with a radiolabeled probe were then compared. HBV DNA was sought in sera and cryoprecipitates that had been kept at -80°C and previously unthawed, using the PCR (Genostics, Abbott Laboratories, Abbott Park, IL). Two sets of primer pairs located in the S (MD06-MD0131 and pre-S genes (MD16-MD191 were used. Viral DNA was purified from 200 FL of serum and subjected to the amplification process for 35 cycles in a DNA thermal cycler (Cetus) as previously described [231. A Southern blot was run and hybridized with a 32P-labeled oligonucleotide probe specific to the amplified region. The sensitivity of the PCR assay was assessed by amplifying serial dilutions of a positive reference known to contain 3 x lo7 molecules of HBV. PCR amplification for Southern blotting was possible with less than 10 HBV DNA molecules. Detection and Characterization

of Cryoglobulins

Cryoglobulins were isolated from the patients’ sera, purified and characterized by immunoblotting at 37°C as previously described [24]. Using this method, only 5/131 (3.8%) healthy blood donors had MC, all of them with a cryoglobulin level lower than 0.03 g/L. In the present study,

96

MIXED

CRYOGLOBULINEMIA

patients were considered to have a significant cryoglobulin if they had a minimum serum cryoglobulin level of 0.05 g/L. After immunochemical analysis, cryoglobulins were classified according to Brouet et al [25]: type I cryoglobulins are single monoclonal immunoglobulins; types II and III are mixed cryoglobulins, composed of different immunoglobulins, with a monoclonal component in type II and only pqlyclonal immunoglobulins in type III. Biochemical Evaluation Hematologic evaluation, serum alanine aminotransferase and aspartate aminotransferase values, alkaline phosphatase level, test for rheumatoid factor and antinuclear Ab, and measurements of serum complement components (C3 and Cd) and total hemolytic complement activity were performed using standard procedures. Circulating immune complexes were detected by polyethyleneglycol precipitation and Clq fixation method. Hematuria and daily proteinuria were also assessed. Liver Histologic Examination A transparietal or transjugular liver biopsy was performed in 30 patients with essential MC and liver test anomalies (23 anti-HCV Ab-positive and 7 anti-HCV Ab-negative patients). Histologic examination of liver biopsies included qualitative and quantitative analyses of lesions. Quantification of periportal necrosis, intralobular necrosis, and portal inflammation and fibrosis generates a histologic activity index for each biopsy, known as Knodell’s score 1261. Clinical Features Clinical features were analyzed retrospectively. For each patient, the following data were recorded: age, sex, presence of arthralgia or arthritis, hypertension (systolic blood pressure greater than 160 mm Hg andlor diastolic blood pressure greater cutaneous involvement than 95 mm Hg), (Raynaud’s phenomenon, purpura, livedo, distal ulcers, or gangrenous changes), signs of clinical hepatic involvement (hepatomegaly, splenomegaly, collateral venous circulation, spider angioma), or peripheral neuropathy. Risk factors for parenteral or community-acquired viral hepatitis were also noted (transfusion, surgery, intravenous drug addiction, and sexual or household contact). Statistical Analysis Statistical analysis used x2 or Fisher’s Exact Test for coEparisons of percentages. Mean quanti-

February

AND

HEPATITIS

C VIRUS

/ CACOUB

ET

AL

TABLE I

Main Clinical Featuresof 63 Patients With Essential Mixed Cryoglobulinemia Anti-HCV AntibodyPositive Patients (n = 33)

Anti-HCV Antibody Negative Patients (n = 30)

p Value

Age at onset

46 + 2

45 t 3

NS

Sex ratio (M/F)

17116

14116

NS

HCV infection risk factors*

11

4

Cutaneous involvemen?

17

5

Peripheral neuropathy

8

18

Arthritisiarthralgia

5

6

NS

Hypertension

5

2

NS

Hepatic involvement*

8

0


to.01 0.004 to.01

ncluaes any 01 tne tollowmg rlsK tactors Tar parenteral or communrty-acqulreo viral nepatms: InsfusIon, surgery intravenous drug addiction, or sexual or householdcontact. Icludes any of thefollowlng: Raynaud’sphenomenon,purpura, livedo, distal ulcers, or gangrenous ranges. icludes any of the following: hepatomegaly,splenomegaly,collateralvenous circulation, or spider angioma.

tative values were compared using the unpaired Student’s t test. Nonparametric analyses were done with the Mann and Whitney U test. Significance was assessed at p = 0.05. All calculated p values are two-tailed. Data are expressed as the means +- standard error of the mean.

RESULTS Using the HCV ELISA, we found anti-HCV Ab in 47/1X (41%) patients with MC: 33163 (52%) essential MC and 14/52 (27%) nonessential MC (p < 0.001); 44 were confirmed by the HCV RIBA, and 3 were indeterminate. Essential MC Anti-HCV Ab were found in 33163 (52%) patients with essential MC by ELISA. Samples from these 33 patients reacted with two (n = 9, 270/o), three (n = 10, 30%), or four (n = 11, 33%) of the four HCV antigens on the RIBA strip. Sera from 3 patients (9%) reacted only with the structural C22-3 protein. Fifty-nine percent, 47%, 63%, and 94% of the RIBA-positive patients had Ab directed against the C5-11, ClOO-3, C33c, and C22-3 HCV proteins, respectively. The 33 anti-HCV Ab-positive patients (Group 1) were compared to the 30 anti-HCV Ab-negative patients (Group 2). The main clinical features of Group 1 and Group 2 patients are listed in Table I. These groups were not different for age, sex ratio, and presence of arthralgia or arthritis, hypertension and renal

1994

The American

Journal

of Medicine

Volume

96

127

MIXED CRYOGLOBULINEMIA AND HEPATITIS C VIRUS / CACOUB ET AL

TABLE II

TABLE IV

Biochemical Data for 63 Patients With Essential Mixed Cryoglobulinemia

Main Virologic Data for 63 Patients With Essential Mixed Cryoglobulinemia

Anti-HCV AntibodyPositive Patients (n = 331

Anti-HCV Antibody Negative Patients (n = 30)

p Value 0.0007

HCV

41 f 10

0.005

HCV ELISA 2 ratio*

76 f 9

0.002

HCV RIBA 2 positivity l&i

AST (IU)

104 f 20

28?

ALT (IU)

110 + 22

Alkaline phosphatase (IU)

130 + 15

5

79 k 3 177,000

87 r3

+ 27,000

Serum creatinine (bmol/L)

94 2 14

Proteinuria (g/d)

0.03 ? 0.02

268,000

rt 20,000

0.0008

83 + 4 0.015

3 Ag 44

NS

3

+ 0.015

Type II cryoglobulinemia

0.35 * 0.07

0.12 2 0.04

HCV RNA in cryoprecipitate+

12/16

0123

0.0000001

NS

HBV

p Value 0.01

8

NS

Circulating immune complexes

8

11

NS

Rheumatoid factor

9

3

NS

Antinuclear antibodies

4

0

NS

28 2 3

44 i 2

0.0001

c3 (g/L)

0.64 i 0.05

0.73 + 0.03

NS

c4 (g/L)

0.20 k 0.02

0.29 k 0.03

February 1994

0.04

The American Journal of Medicine

-

i!

NS

12

Total hemolytic complement (CHsolmL)

-

0.00001

Immunologic Data for 63 Patients With Essential Mixed Cryoglobulinemia

Cryoglobulin serum level (g/L)

0.0001

O/23

TABLE Ill

Anti-HCV AntibodyNegative Patients (n = 30)

0.16 + 0.02

p Value

lo/16

involvement. Group 1 patients had cutaneous involvement and signs of clinical hepatic involvement more frequently and peripheral neuropathy less frequently than Group 2. Risk factors for HCV transmission were found in 11 Group 1 patients, as follows: history of transfusion (n = 6) or surgery (n = 4), and sexual contact (n = 1). These risk factors were more common than in Group 2 patients. Biochemical data are given in Table II. Liver function test abnormalities were more marked in Group 1 patients. Liver function tests were abnormal in 20/33 (61%) and 4/30 (13%) Group 1 and

Anti-HCV Antibody Positive Patients (n = 33)

9

Anti-HCV AntibodyNegative Patients (n = 30)

HCV RNA in serum+

T = serum aspartateaminotransferase; ALT = serum alanineaminotransferase.

128

5.40 i 0.15

2 Ag

Prothrombin time (%I Platelets (/mm3)

An&HCV Antibody Positive Patients (n = 33)

HBs Ag

1

1

NS

HBs Ab

8

2

NS

HBcAb

13

4

0

0

NS

0112

0112

NS

HBV DNA in serum+ HBV DNA in cryoprecipitatef

to.05

LISA ratio: optical density/cutoff. y polymerase chain reaction.

Group 2 patients, respectively (p = 0.0001). Neither group had renal involvement. Table III summarizes the main immunologic data. Group 1 patients had significantly higher serum cryoglobulin levels. Type II cryoglobulins were found in the same, proportion of Group 1 and Group 2 patients, but the former contained monoclonal IgM with K light chains more often (11 versus 3; p < 0.02), and IgG A less often (0 versus 4; p = 0.04), whereas there was no difference in the IgM X component (1 versus 1). Sera from Group 1 patients usually showed decreased total hemolytic complement activity and C4 levels. Rheumatoid factor and antinuclear Ab were rarely detected in both groups. No correlation was found between the serum cryoglobulin level and the HCV ELISA ratio, or between the serum cryoglobulin level and the serum alanine aminotransferase level (data not shown). Serum HBs Ag was observed in only one patient from each group (Table Iv>. The prevalence of anti-HBc Ab was higher in Group 1 patients than in Group 2. The prevalence, in an isolated manner, of one HBV serum marker (HBs Ag, HBs Ab, or HBc Ab) was higher in Group 1 patients than in Group 2 (16/33 = 48% versus 5130 = 17%;

Volume 96

MIXED

CRYOGLOBULINEMIA

p = 0.04). HBV DNA sequences were not detected in any of the sera or cryoprecipitates tested. The sera and cryoprecipitates of 39 unselected patients with essential MC were tested for the presence of HCV sequences by means of PCR (Table 4). HCV RNA sequences were detected in lo/16 (63%) sera and 12/16 (75%) cryoprecipitates from Group 1 patients. HCV RNA was not found in either the sera or the cryoprecipitates of 23 Group 2 patients. In two control groups of patients without MC, HCV RNA sequences were identified in 40/50 (80%) sera from patients with chronic hepatitis C, whereas HCV RNA was never found in 100 sera from anti-HCV Ab-negative patients. In the six cases of MC with anti-HCV Ab in whose sera and cryoprecipitates were studied in parallel, the concentration of anti-HCV or antiHBc Ab (expressed as optical density): was not significantly higher in cryoprecipitates than in whole serum or supernatants. With quantitative PCR, HCV RNA was found to be more concentrated in cryoprecipitates than in supernatants in the three samples examined (20 times more in cryoprecipitates than in supernatants in 2 cases, and 100 times more in 1 case). Liver histologic examination was performed in 23 Group 1 patients and revealed a spectrum of lesions ranging from chronic active hepatitis (n = 10,43%) to partial or frank cirrhosis (n = 15, 65%). The liver biopsies obtained from 7 Group 2 patients showed micronodular cirrhosis (n = 4, 57%) or chronic active hepatitis (n = 3, 43%). Knodell’s score was comparable between Group 1 and Group 2 patients, 8.9 & 2.2 versus 8.0 +- 2.0, respectively. Nonessential MC In the 52 patients with nonessential MC, antiHCV Ab were found in 14 patients (27%) (Group 3) with a mean ELISA ratio of 5.10 * 0.37, and they were undetectable in 38 patients (73%) (Group 4). Samples from Group 3 patients reacted with two (n = 4, 29%), three (n = 4, 29%), or four (n = 2, 14%) of the four HCV antigens on the RIBA strip. Sera from four patients (29%) reacted only with the structural C22-3 protein. Twenty-nine percent, 36%, 71%, and 100% of,Group 3 patients had Ab directed against the C5-l-1, ClOO-3, C33c, and C22-3 HCV proteins, respectively. Considering the diagnosis of underlying disease, anti-HCV Ab were found in 9/33 (27%) patients with autoimmune or related disorders (polyarteritis nodosa n = 2, rheumatoid arthritis, systemic lupus erythematosus, autoimmune thrombocytope-

February

AND

HEPATITIS

C VIRUS

/ CACOUB

ET AL

nit purpura, cerebral angiitis, sarcoidosis, primary biliary’cirrhosis, primary Sjiigren’s syndrome), 3/11 (27%) patients with nonviral infection (malaria + leishmaniasis, malaria + syphilis, rickettsialpox), and Z/l0 (20%) patients with malignant hematologic disorders (chronic myelogenous leukemia, Waldenstrom’s disease). Group 3 and Group 4 patients were comparable for age (45 5 3 versus 41 & 3 years), male-tofemale sex ratio (618 versus 21/17), and cutaneous (6 versus 19), articular (5 versus 14), renal (0 versus 2), and neurologic (7 versus 14) signs. Group 3 patients had systemic hypertension more frequently than Group 4 (5/14 = 36% versus 4138 = 11%; p <0.005). Risk factors for HCV transmission were identified in a comparable proportion of patients in the two groups (7114 Group 3 versus 14/38 Group 4 patients). Most hematologic and immunologic data did not differ between Group 3 and Group 4 patients. Serum aspartate aminotransferase was 68 + 18 versus 46 -t 11 II-J; serum alanine aminotransferase, 74 * 18 versus 47 + 17 IU; alkaline phosphatase, 83 + 9 versus 100 ? 15 IU; platelet count, 202,000 + 27,000 versus 282,000 + 21,000/mm3; total hemolytic complement activity, 33 2 6 versus 36 + 2 CHso/mL; C4 levels, 0.23 + 0.04 versus 0.22 + 0.03 g/L. Abnormal liver function tests were noted more often, however, in Group 3 (43%) than in Group 4 patients (34%) (p < 0.001). There was a trend towards higher serum cryoglobulin levels in Group 3 patients (0.86 -+ 0.60 versus 0.25 + 0.09), but this difference was not significant. Group 3 patients had type II cryoglobulinemia more often (7114 = 50% versus 10138 = 26%; p
COMMENTS Using second-generation HCV tests, we found a high prevalence (41%) of anti-HCV Ab in 115 consecutive unselected patients with MC. This anti-HCV Ab prevalence was even higher (52%) in the 63 patients defined as having essential MC in the absence of a well-defined connective-tissue disease, malignant hematologic disorder, or obvi-

1994

The American

Journal

of Medicine

Volume

96

129

MIXED

CRYOGLOBULINEMIA

AND

HEPATITIS

C VIRUS

/ CACOUB

i ous nonviral infectious process. In another study, Lunel et al reported a very high MC prevalence in patients with chronic HCV hepatitis never treated with interferon-alpha (40/75 = 53%) [27]. This high prevalence of MC was not due to an excessive sensitivity of the method of cryoglobulin assessment [241, because cryoglobulins of a 0.05 g/l minimum serum level were never detected in healthy blood donors, a finding that is similar to previously published results [281. The fact that results obtained by ELISA were confirmed by the RIBA excludes the possibility of false-positive results due to hypergammaglobulinemia 1291 or the presence of cryoglobulinemia itself 1301. The high prevalence of HCV RNA found in the sera and/or cryoprecipitates of patients with anti-HCV Ab and MC (79%) also sustains the hypothesis that HCV has a direct role in the production of MC. In MC patients, the hepatotropic antigen(s) that can trigger the production of antibodies that can later form immune complexes has been sought for many years. The association of MC and HCV infection was first noted using a first-generation test in 10 patients with type II MC, 3 of whom had anti-HCV Ab 1151. Since then, many case reports or congress abstracts have been published, but only a few prospective studies with a large series of unselected MC patients have analyzed the prevalence of anti-HCV Ab and HCV RNA, both in sera and cryoprecipitates. Previously published studies contained numerous biases and/or incomplete data: the MC population was highly selected and composed of only Italian patients in whom the prevalence of HCV serum markers is particularly high (twice that of the French population) 119,201; they included few MC patients (less than 25 patients) [18,31-331; neither the RNA prevalence 118,311 nor the presence of HCV markers in cryoprecipitates was analyzed [18,19,311. The design of our study was original because we conducted a prospective study on a large population of French patients with essential MC seen in different departments of our university hospital. We evaluated the prevalence of both anti-HCV Ab and HCV RNA in sera and cryoprecipitates. We also compared the clinical, biochemical and virologic data of essential MC patients with or without anti-HCV Ab; a comparable investigation has not been reported thus far. We found anti-HCV Ab in about 50% of the unselected essential MC patients using the secondgeneration tests. HCV seems to be directly involved in MC pathogenesis because HCV RNA sequences were detected in 64% of the sera and 79% of the cryoprecipitates from patients with MC

130

February

1994

The American

Journal

of Medicine

Volume

ET AL

and anti-HCV Ab. Using quantitative PCR, moreover, we found a selective concentration (from 20 to 100 times) of HCV RNA in the cryoprecipitate, as recently reported 1331. From a clinical point of view, compared to our anti-HCV Ab-negative patients, our anti-HCV Ab-positive patients more often had cryoglobulin-related cutaneous involvement and clinical and biochemical hepatic signs that were not noted in previous studies. The particularly important role of HCV in the development of MC may also explain a 2- to 3-fold increased frequency of MC in HCV chronic hepatitis compared to HBV or non-A, non-B, non-C chronic liver disease [27]. It is noteworthy that the geographic distributions of MC and HCV infection are quite similar. Mixed cryoglobulins and HCV infection are relatively uncommon in northern Europe and North America but seem to be more common in southern Europe 1341. A possible explanation for the HCV-MC association could be an antigenic cross-reaction between HCV and the liver, pai-ticularly a damaged liver, with the production of Ab cross-reacting with both antigens. The possibility of a cross-reactive Ab is strengthened by the finding that HCV infection induces an Ab response to an antigen apparently encoded by the host genome [351. In the present study and that by Ferri et al 1191, the presence of HCV RNA was not correlated with the presence or absence of biopsyproven liver involvement, suggesting an extrahepatic pathogenetic pathway of HCV for excessive cryoglobulin production. The high prevalence of HCV serum markers in our patients may also reflect HCV exposure as a consequence rather than a cause of illness. The patients with essential MC, for example, may have required more frequent hospitalizations or transfusions. This hypothesis seems unconvincing, however, because patients with nonessential MC have a significantly lower prevalence of HCV serum markers than patients with essential MC (27% versus 52%; p
96

MIXED

CRYDGLDBULINEMIA

Other hepatotropic antigens that can induce the synthesis of antibodies that can then form immune complexes have been sought for many years. Levo et al [7] first implicated HBV as a causative agent of essential MC, and subsequent studies claimed a high prevalence of HBV surface markers in both sera and cryoprecipitates of patients with essential MC [4,6,8,38]. Numerous studies using more recent tests for the detection of HBV markers and HBV DNA, however, failed to validate these observations [5,9-11,391. More recently, serologic evidence of HBV infection, defined only by serum anti-HBc positivity, was reported in 50% to 75% of patients with essential MC and anti-HCV-Ab, with a significant association between chronic liver disease and both HCV and HBV infections [l&191. In the present study, the presence of HBs Ag or anti-HBs Ab was noted in 17% of the anti-HCV Ab-negative patients and in 48% of anti-HCV Ab-positive patients (p < 0.05). The higher prevalence of HBV serum markers in the HCV Abpositive population is well-known and not surprising because these viruses have common modes of transmission 1391. None of our anti-HCV Abpositive patients,. however, had detectable HBV DNA in their sera and/or cryoprecipitates, indicating the probable absence of a causative role for HBV in liver function test abnormalities and/or MC in these patients. In conclusion, about 50% of the essential MC patients had anti-HCV Ab and had more severe cryoglobulinemia-associated clinical and biological signs. HCV RNA sequences were found in the large majority of sera and cryoprecipitates from patients with essential MC and anti-HCV Ab, and HCV RNA appears to be selectively concentrated in cryoprecipitates. These results suggest that HCV may now be considered as an agent responsible for cryoglobulinemia and that, in this context, many cases of MC diagnosed as essential may indeed be the consequence of HCV infection and thus nonessential. REFERENCES 1. Meltzer M, Franklin EC. Cryoglobulinemia: a study of twenty-nine patients. Am J Med 1966; 40: 828-36. 2. Meltzer M, Elias K, McCluskey RT, Cooper N, Franklin EC. Cryoglobulinemia-a clinical and laboratory study. Am J Med 1966; 40: 837-56. 3. Levo Y, Gorevic PD, Hanna MD, Kassab HJ, Tobias H, Franklin EC. Liver involvement in the syndrome of mixed cryoglobulinemia. Ann Intern Med 1977;87:287-92. 4. Gorevic PD, Kassab HJ, Levo Y, et al. Mixed cryoglobulinemia: clinical aspects and long-term follow-up of 40 patients. Am J Med 1980; 69: 287-308. 5. Zarski JP, Rougier D, Aubert H, et al. Association cryoglobuline et maladie hepatique: frequence, nature et caracteres immuno-chimiques de la cryoglobulinemie. Gastroenterol Clin Biol 1984; 8: 845-50.

February

6. Garcia-Bragado Cryoglobulinemie Nouv Presse Med 7. Levo Y, Gorevic

AND

HEPATITIS

C VIRUS

/ CACOUB

ET

AL

F, Vilardell M, Fonollosa V, Ruibal A, Gallart T, Cuxart A. mixte essentielle: relations avec le virus de I’hepatite B. 1981; 10: 2955-7. PD, Kassab HJ, Zucker-Franklin D, Franklin EC. Associa-

tion between hepatitis B virus and essential mixed cryoglobulin. N Engl J Med 1977; 296: 1501-4. 8. Bombardieri S, Ferri C, Di Munno 0, Pasero G. Liver involvement in essential mixed cryoglobulinemia. RicCiin Lab 1979; 9: 361-8. 9. Dienstag JL, Wands JR, Benson GD, lsselbacher KJ. Hepatitis B and essential mixed cryoglobulinemia. N Engl J Med 1977; 297: 946 (letter). 10. Popp JW, Dienstag JL, Wands JR, Bloch KJ. Essential mixed cryoglobulinemia without evidence for hepatitis B virus infection. Ann Intern Med 1984; 92: 379-83. 11. Garcia-Bragado F, Biosca M, Villar M, De La Figuera M, Rodrigo MJ, Vilardell M. Maladies hepatiques et cryoglobulinemie mixte. Gastroenterol Clin Biol 1984; 9: 456-7. 12. Galli M, Careddu F, D’Armino A, Monti G, Messina K, lnvernizzi F. Hepatitis B virus and essential mixed cryoglobulinemia. Lancet 1980; 1: 1093 (letter). 13. Alter MJ, Sampliner R. Hepatitis C: and miles togo beforewesleep. N Engl J Med 1989; 321: 1538-9. 14. Durand JM, Lefevre P, Harle JR, Boucrat J, Vivitski L, Soubeyrand J. Cutaneous vasculitis and cryoglobulinaemia type II associated with hepatitis C virus infection. Lancet 1991; 337: 499-500. 15. Pascual M, Perrin L, Giostra E, Schifferli JA. Hepatitis C virus in patients with cryoglobulinemia type II. J Infect Dis 1990; 162: 569 (letter). 16. Harle JR, Disdier P, Durand JM, et al. Cryoglobulinemie I’infection par le virus de I’hepatite C. Dix cas. Presse Med (letter). 17. Taillan B, Ferrari P, Fuzibet JG, Ferrari E, Quaranta Cryoglobulinemie mixte au tours de I’infection par le virus Interetde I’interferon. Presse Med 1991; 20: 1392-3.

mixte au tours de 1991; 26: 1233 JF, Dujardin de I’hepatite

P. C.

18. Casato M, Pucillo LP, Lagana B, Taliani G, Goffredo F, Bonomo L. Cryoglobulinaemia and hepatitis Cvirus. Lancet 1991; 337: 1047-50. 19. Ferri C, Greco F, Longobardo G, et al. Antibodies to hepatitis C virus in patients with mixed cryoglobulinemia. Arthritis Rheum 1991; 34: 1606-10. 20. Misiani R, Bellavita P, Fenili D, eta/. Hepatitis Cvirus infection in patients with essential mixed cryoglobulinemia. Ann Intern Med 1992; 117: 574-7. 21. Cacoub P, Musset L, Lunel-Fabiani F, eta/. Hepatitis C virus and essential mixed cryoglobulinemia. Br J Rheumatol 1993; 32: 685-92. 22. Weiner AJ, Kuo G, Bradley DW, et al. Detection of hepatitis C viral sequences in non-A, non-B hepatitis. Lancet 1990; 335: 1-3. 23. Thiers V, Nakajima E, Kremsdori D, et al. Transmission of hepatitis B from hepatitis B-seronegative subjects. Lancet 1988; 2: 1273-6. 24. Musset L, Diemert M-C, Taibi F, eta/. Characterization of cryoglobulins by immunoblotting. Clin Chem 1992; 38: 798-802. 25. Brouet JC, Clauvel JP, Danon F, Klein M, Seligman M. Biologic and clinical significanceof cryoglobulins. Am J Med 1974; 57: 775-88. 26. Knodeli RG, lshak KG, Black WC, et a/. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981; 1: 431-5. 27. Lunel F, Mosset L, Cacoub P, et al. Cryoglobulinemia in chronic liver diseases: role of hepatitis C virus and liver damage. Gastroenterology (in press). 28. Cream JJ. Clinical and immunological aspects of cutaneous vasculitis. Q J Med 1976; 178: 255-76. 29. McFarlane IG, Smith HM, Johnson PJ, Bray GP, Vergani D, Williams R. Hepatitis C virus antibodies in chronic active hepatitis: pathogenetic factor or false positive result? Lancet 1990; 335: 754-7. 30. Theilmann L, Blazek M, Goeser T, Gmelin K, Kommerell B, Fiehn W. False positive anti-HCV tests in rheumatoid arthritis. Lancet 1990; 2: 1346 (letter). 31. Dammaco F, Sansonno D. Antibodies to hepatitis C virus in essential mixed cryoglobulinemia. Clin Exp lmmunol 1992; 87: 352-6. 32. Pechere-Bertschi A, Perrin L, de Saussure P, Widmann JJ, Giostra E, Schifferli JA. Hepatitis C: a possible etiology for cryoglobulinemia type II, Clin Exp lmmunol 1992; 89: 419-22. 33. Agnello V, Chung RT, Kaplan LM. A role for hepatitis C virus infection in type II cryoglobulinemia. N Engl J Med 1992; 327: 1490-5.

1994

The American

Journal

of Medicine

Volume

96

131

MIXED

CRYOGLOBULINEMIA

34. Phillips PE, Dougherty

AND

RM. Hepatitis

HEPATITIS

C VIRUS

/ CACOUB

C virus and mixed cryoglobulinemra.

Clin Exp Rheumatol 1991; 9: 551-5. 35. Mishiro S, Hoshi Y, Takeda K, et al. Non-A, non-B hepatitis specific antibodies directed at host derived epitope: implication for an autoimmune process, Lancet 1990; 336: 1400-3. 36. Alter HJ, Purcell RH, Shih JW, et al. Detection of antibody to hepatitis C virus in prospectively followed transfusion recipients with acute and chronic non-A, non-B hepatitis. N Engl J Med 1989; 321: 1494-1500.

132

February

1994

The American

Journal

of Medicine

Volume

ET AL 37. Columbo M, Oldani S, Donato MF, eta/. A multicenter prospectivestudyof post-transfusion hepatitis in Milan. Hepatology 1987; 7: 702-12. 38. Realdi G, Alberti A, Rigoli A, Tremolada F. Immune-complexes Australia antigen in cryoglobulinemic sera. Z lmmunitaetsforsch 1974; 114-26. 39. Galli M. Cryoglobuiinemia Lancet 1991; 338: 758-9.

96

and serological

markers

of hepatitis

and 147:

viruses.