Hepatitis in patients with acute nonlymphocytic leukemia

Hepatitis in Patients with Acute Nonlymphocytic Leukemia

JAMES C. WADE, M.D. MICHAEL GAFFEY, MS. PETER H. WIERNIK, M.D. STEPHEN C. SCHIMPFF, M.D. CHARLES A. SCHIFFER, M.D. MARGARET WESLEY, Ph.D. Baltimore,

Marylend

JAY H. HOOFNAGLE, M.D. Bethesda, Maryland

From the University of Maryland Cancer Center, University of Maryland School of Medicine, Raltimore. Marvland. and the Liver Disease Section, Dbestive Diseases Branch, National Institute of AGhritis, Metabolism and Digestive Diseases, Bethesda, Maryland. This work in part was previously submitted and published in abstract form in the Proceedings of the 16th Annual Meeting of the American Society of Clinical Oncology, May 1960, San Diego, California. Requests for reprints should be addressed to Dr. James C. Wade, University of Maryland Cancer Center, University of Maryland Hospital, 22 South Greene Street, Baltimore, Maryland 2 120 1. Manuscript accepted January 16, 1963.

Three consecutive groups (University of Maryland Cancer Center protocols 7110, 7405, and 7802) of patients with acute nonlymphocytlc leukemia who achieved a complete hematologlc remission with similar antileukemic therapy were reviewed for the development of hepatitis. Ninety-four (73 percent) experienced viral hepatitis; eight had type B hepatitis and 88 had non-A/non-B hepatitis. The hepatitis was mild in all patients. Hepatitis secondary to cytomegalovirus, herpes simplex virus, Epstein-Barr virus, or Toxoplasma gondil was not observed. Antibody to type A hepatitis was common, but acute infection could not be substantiated. All cases of type B hepatltls in which the surface antfgen could be serotyped were found to have the less frequently observed ayw marker, suggesting a common donor as the source of infection. The medlan duration of complete remission was longer (p = 0.03) for patients in Group II (protocol 7405) who contracted hepatltls (247 days) compared with patients without hepatitis (125 days). Median overall survival was also longer (p = 0.01) for these patients in whom hepatltis developed (872 days versus 372 days, respectively). No prolongation of complete remission duration or survival could be demonstrated for patlents from Group I (protocol 7110) or Group Ill (protocol 7802) who contracted hepatitis. In patients with hepatitis, the height of transaminase serum bilirubin levels or duration of abnormal results of liver function tests did not correlate with the duration of complete remission or survival. Hepatitis, a common infection in those patients with acute nonlymphocytic leukemia who undergo induction therapy, had an inconsistent effect on the duration of complete remission interval and overall survival. Viral hepatitis is a serious post-transfusion infectious complication and, despite the routine screening of blood donors for type B hepatitis surface antigen, it still develops in approximately 7 percent of blood recipients [l-4]. Type B hepatitis accounts for only 10 to 15 percent of the cases [3,4], with non-A/non-B hepatitis accounting for the remainder. The episode rate for post-transfusion hepatitis in uatients who receive blood from commercial donors shows a dose&elated incidence [ 1,4-71, although the hepatitis episode rate in recipients of volunteer donor blood appears to be largely independent of the number of transfusions [ 1,2,6,7]. Patients with acute nonlymphocytic leukemia receive multiple red blood cell and platelet concentrate transfusions during remission induction and subsequent intensive maintenance therapy. Levels of liver enzymes frequently are elevated in such patients, but the exact cause is often clouded by the recent administration of chemotherapeutic agents, other potential hepato-

September 1983

The American Journal of Medicine

Volume 75

413

‘-VAT’%

Ah”[3 AC’L’TE ‘LEUKEMIA-WADE

ET AL

t’oxins, bacterial and fungal infections, and leukemic infiltration of the liver. A retrospective evaluation from Alabama of hepatitis in 50 patients with acute nonlymphocytic leukemia found that, contrary to the supposition that hepatitis would be detrimental, the occurrence of hepatitis improved survival [8]. We have had a continuing interest in hepatitis and its frequent occurrence during or shortly after the initial remission induction phase of acute nonlymphocytic leukemia [Q]. In this report, the incidence, cause, course, and implications of presumed viral hepatitis are retrospectively evaluated in three groups of consecu-

tively treated patients with acute nonlymphocytic kemia.

leu-

PATIENTS AND METHODS Group I. Sixty-four consecutive patients with acute nonlymphocytic leukemia were entered onto a single-treatment protocol at the University of Maryland Cancer Center (UMCC) beginning in1971 (UMCC protocol 7110). Remission induction therapy consisted of either daunorubicin alone (60 mg/m2 daily for three days) or daunorubicin, cytarabine, Ei-thioguanine, and pyrimethamine. Remission maintenance therapy consisted of two consolidation courses, using the successful remission induction regimen at half dose, and then all patients received monthly cyclophosphamide (1 g/m2 intravenously) plus guanazole (25 g/m2 intravenously) until relapse or death [lOI. Group II. Sixty-seven consecutive patients with acute nonlymphocytic leukemia were entered onto a single-treatment protocol (UMCC protocol 7405) of combined therapy with daunorubicin and cytarabine beginning in 1974. Daunorubicin,

45

mg/m2 per day for three days by rapid intra-

venous infusion, and concomitant cytarabine, 100 mg/m2 per day for seven days by continuous intravenous infusion, were the remission induction agents. Patients were randomized to receive or not to receive levamisole, 100 mg orally every 12 hours, three days per week. Once bone marrow remission was obtained, two courses of daunorubicin, 45 mg/m2 per day for two days, and cytarabine, 100 mg/m2 per day by continuous infusion for five days, were given as consolidation therapy. If complete remission persisted, patients then received three monthly infusions of methotrexate at escalating doses (50 mg/kg, 100 mglkg, and 150 mg/kg) with accompanying folinic acid reversal (cytoreduction therapy). Intensive maintenance therapy (utilizing 6-thioguanine orally plus cytarabine intravenously given on a daily basis until the bone marrow was hypoplastic) was then administered every three months to those patients who remained in complete remission, and was continued until the patient relapsed or died from other causes. Group III. Seventy-seven patients with acute nonlymphocytic leukemia were entered onto a third-treatment protocol of combined therapy with daunorubicin and cytarabine (UMCC protocol 7802). The induction phase was identical to protocol 7405. Once bone marrow remission was obtained, intensive maintenance chemotherapy was initiated (with oral 6-thioguanine plus subcutaneous cytarabine given on a daily basis

414

September 1993

The American Journal of Medicine

until bone marrow aplasia, and then repeated every 10 weeks until the patient relapsed or died from other causes). All patients with a complete remission underwent randomization to receive either: (1) intensive maintenance chemotherapy alone or (2) intensive maintenance chemotherapy plus neuraminidase-treated allogenic myeloblasts or (3) intensive maintenance chemotherapy plus splenectomy. Because new patients are still being entered into this latter study, we chose to analyze only those patients entered as of June 30, 1980, so as to ensure a minimum of one-year follow-up as of the final evaluation date of June 30, 1981. All patients were attended by a uniform cohort of medical and paramedical personnel in a single hospital ward and adjacent outpatient facility and were followed at this institution for the remainder of their lives. Once patients had achieved a complete remission, bone marrow examinations to assess remission status were performed on a regular schedule prior to each course of maintenance chemotherapy, and at any other time when clinical findings suggested the appearance of leukemic relapse, or marrow hypofunction. During periods of myelosuppression, patients were supported with routine infection prevention and therapy techniques, and red blood cell and prophylactic platelet transfusions for platelet counts of less than 20,000/~1. Therapeutic granulocyte transfusions were administered only to patients who had documented bacterial infections that were unresponsive to appropriate antibiotics. Although these approaches changed over the lo-year course of this study, the approaches utilized at any one time were applied equally to all patients. Blood chemistry determinations were performed using standardized automated methods (Technicon SMAC; Technicon Instrument Corporation, Tarrytown, New York) at least twice weekly during remission induction and at least monthly while the hematologic remission was maintained. Serum glutamic oxalacetic transaminase levels were determined by an automated method. Serial serum specimens that had been previously collected during the induction period from patients in Group II (UMCC protocol 7405) and frozen at -70°C were retrieved and tested (J.H.H.) for hepatitis B surface antigen by radioimmunoassay (Austria II, Abbott Laboratories, North Chicago, Illinois) [ 111, antibody to hepatitis B surface antigen by radioimmunoassay (Ausab, Abbott Laboratories) [ 121, and antibody to hepatitis B core antigen by a solid-phase, competitive inhibition radioimmunoassay (Corab, Abbott Laboratories) [ 131. Serum samples that were found to be hepatitis B e surface antigen-positive were serotyped when possible, and the presence of hepatitis B e antigen was determined by immunodiffusion and radioimmunoassay. Antibody to hepatitis type A was determined by radioimmunoassay in all baseline, acute, and convalescent serum samples [ 141. All possible seroconversions were evaluated for IgM antibody to hepatitis A virus [ 151. Aliquots of the same serial serum samples were sent to the Maryland State Department of Health where complement fixation titers for cytomegalovirus, Epstein-Barr virus, and herpes simplex, as well as dye test for Toxoplasma, were performed through the courtesy of J. M. Joseph, Ph.D. The medical record was reviewed for all patients in the three study groups. The period of hepatitis risk was defined as beginning with the day of arrival at the University of

Volume 75

NEPATlTrsAND ACUTELEUKEMIA---WADE

Maryland Cancer Center for remission induction chemotherapy and continuing for 160 days after the date of administration of the last remission induction-associated blood product transfusion. Hepatitis was defined, utilizing standard criteria, as an elevation in the transaminase level of two and a half times normal (5 to 20 mU/ml) on two consecutive determinations at least five days apart [ 16,171. lcteric hepatitis was defined as a peak serum bilirubin level of 2 mg/dl or greater [ 16,171. Patients with elevated transaminase values at the time of initial induction were not included unless the levels of liver enzymes returned to normal for at least IO days before subsequent exacerbation. Liver biopsy was performed at the discretion of the patient’s primary physician. The biopsy specimens were retrospectively reviewed by the University of Maryland Cancer Center pathologist, John Sutherland, M.D., without knowledge of the patient’s medical record. Results of liver biopsy, when available, were used to distinguish between infectious or noninfectious hepatitis. Drugrelated heparoroxiciry was defined as compatible histopathologic changes by liver biopsy and/or recurrent elevations in the serum glutamic oxalacetic transaminase level that occurred with readministration of the same drug. Patients believed to have drug-related hepatotoxicity were included in the “no hepatitis” group for the purpose of analysis. Viral hepatitis was categorized on the basis of serologic studies. Type 13 hepariris occurred when elevation of the transaminase level was accompanied by the presence of hepatitis l3 surface antigen or when type B hepatitis seroconversion occurred. Seroconversion was defined as the appearance of antibody to hepatitis B surface antigen (at a radioimmunoassay ratio of 2: 1 or greater) and antibody to hepatitis B core antigen (inhibition, at a level of greater than 50 percent, when compared with positive and negative controls) [ 181 in two consecutively drawn serum specimens. IgM antibody directed against core antigen was assayed in all patients observed to have had a seroconversion for type B hepatitis. Type A hepatitis was diagnosed when the transaminase criterion was met, seroconversion occurred, and type A hepatitis IgM antibody was detected. Hepatitis secondary to cytomegalovirus, Epstein-Barr virus, and herpes simplex required a fourfold rise in titer, whereas Toxoplasma was incriminated only when an IgM dye test titer of 1: 16 or greater was found to accompany the rise in the transaminase level. Only when results of all serologic evaluations were negative and acute infection-related and drug-related elevations in the transaminase level were excluded was hepatitis classified as non-A/non-B. Survival was recorded from the first day of induction chemotherapy to death. Complete remission duration was calculated from the first day a complete remission was documented to the day that bone marrow examination revealed leukemic relapse. Hepariris incubation period has been defined as the time from the first blood product transfusion following arrival at our center (i.e., maximum incubation period) to detection of hepatitis. All blood components administered to these patients were derived from similar sources. Red blood cells and fresh frozen plasma were obtained through the University of Maryland Blood Bank and the American Red Cross from volunteer donors who had been screened for hepatitis B surface anti-

ET AL

gen. Granulocytes were obtained from ABC-matched donors, most of whom were patient family members. Most tranfused platelets were obtained from a paid commercial donor pool, which for UMCC protocol 7405 was screened periodically for history, physical examination, serum bilirubin level, and at each donation for hepatitis B surface antigen by radioimmunoassay. During the period since initiation of UMCC protocol 7802, all platelet donors have also been screened monthly for serum glutamic oxalacetic transaminase levels. Statistical evaluation of remission duration and total survival utilized a time-dependent covariable to account for the possibility that a relationship may exist between prolonged survival and a higher probability of hepatitis developing the longer the patient lived. The conditionality principle presented by Cox is elemental to this analysis and states that the probability of a person responding at a given time is his hazard function at that moment divided by the sum of all individual hazard functions noted to be at risk of responding simultaneously [ 191. A modified life table form was used. Patients are initially divided into two groups, those with hepatitis and those without. After induction chemotherapy, those in the nonhepatitis group may be transferred out of their group (tallied as a “loss”) and placed on the first day of illness into the hepatitis group (tallied as an “addition”). For analysis, the state of not having hepatitis is transient, and losses due to such transfer will be distinct from losses to observation (patient death). Following tabulation of the life table, the data were analyzed according to the method of Mantel and Byar [20]. The remaining analyses utilized a Fisher exact test to compare proportions and Student t test to compare means. RESULTS

Group I. Sixty-four

consecutive patients from UMCC protocol 7110 received induction therapy; 36 (56 percent) achieved a complete remission. The median remission duration was 188 days (mean = 186.2 days, range 27 to 590 days), which did not differ significantly between induction regimens. Hepatitis developed in 33 of the 36 patients who achieved a complete remission. Three of these patients were believed to have drug-related hepatotoxicity, while the fourth had changes in the liver enzyme level believed due to leukemia infiltration of the liver. Thus, viral hepatitis occurred in 29 (81 percent) of the 36 patients (Table I). Peak serum glutamic oxalacetic transaminase levels were high (mean 1,275, median 432, range 110 to 6,500 mu/ml) (Table II). Two patients had positive results of an assay for hepatitis B surface antigen, but it is difficult to further categorize these cases of hepatitis since the radioimmunoassay for type B hepatitis was not readily available at that time in our hospital and only a few patients were tested. Group II. Sixty-seven patients on UMCC protocol 7405 received induction chemotherapy with daunorubicin and cytarabine; 42 of these patients (63 percent) obtained a complete remission, with a median remission duration

September 1983

The American Journal of Medicine

Volume 75

415

TABLIE I

Characteristics of Patients with Acute Nonlymphocytic Leukemia Who Achieved a Complete Remission Protocol 7405 Group Ila Group Ilb (Hepatitis) (No Hepatffls)

Protocol 7 110 Group lb Group la (Hepatitis) (No Hepatitis) Total patients Sex (male/female) Mean age (yr) Mean number of remission induction courses Mean number of blood products given* Red blood cells Platelets Others+

Protocol 7802 Group llla Group lllb (Hepatitis) (No Hepatitis)

29 18111 48 (19-79) NA

7 512 39 (15-64) NA

33 16717 42 (16-68) 1.4 (l-3)

9 376 42 (16-69) 1.1 (l-2)

32 13119 43 (14-67) 1.2 (l-2)

19 8111 40 (14-72) 1.1 (l-3)

61 22 (O-59) 38 (4-75) 1 (O-17)

66 25 (10-44) 39 (7-75) 2 (O-6)

72 12 (O-36) 58 (9-143) 2 (O-21)

74 16 (6-35) 56 (18-94) 2 (O-19)

86 11 (l-39) 72 (4-l 14) 3 (O-14)

81 17 (4-32) 61 (12-84) 3 (O-21)

Units administered. t Granulocytes, fresh frozen plasma. ( ) = range; NA = not available. l

TABLE II

Viral Hepatitis: Type, Diagnosis, and Characteristics

TyPs 8 Hepatitis Number of patients Serologic test results Type B hepatitis Hepatitis B surface antigen-positive Seroconversion Mean blood product exposures Red blood cells Platelet concentrates Hepatitis characteristics (median) Maximum incubation period (days) Duration of abnormal transaminase level (days) Peak transaminase level (mu/ml) Peak bilirubin level (mg/dl)

Protocol 7110 Non-A/Non-B Hepatitis

2

27

2

0

50 6 44

Protocol 7405 TyPs 8 Non-A/Non-B Hepatitis Hepatitis 4

61 23 (O-59) 38 (4-75)

29

o+

Protocol 7802 Type 6 Non-A/Non-B Hepatitis Hepatitis 2

3’ 1 80 12 (5-22) 68 (9-127)

0 67 12 (O-36) 55 (18-143)

89 13 (O-25) 76 (73-79)

2

104 4s

86 (13-134) 85 (5-315)

148 (102-170) 142 (28-212)

77 (19-143) 114 (10-518)

94 (71-119) 212 (34-390)

67 (10-133) 78 (12-740)

310 0.8

1,275 (llO-6,500) 4.6 (0.6-17.0)

142 (83-855) 1.4 (1.3-3.5)

393 (58-2.200) 4.0 (0.7-19.1)

140 (100-180) 0.7 (0.8-0.8)

346 (77-2,620) 1.6 (0.4-2.0)

Hepatitis B surface antigen, all type ayw. t Hepatitis B surface antigen-positive hepatitis later developed in three of 29 patients with non-A/non-B hepatitis and a patient who was free from hepatitis during the 160day study period. Three of the four had typeable hepatitis B surface antigen; each had the ayw serotype. f One of 30 patients with non-A/non-B hepatitis became positive for hepatitis B surface antigen after 160day study period. 5 Units. ( ) = range. l

of 183 days (range, 6 to 2,258-I) and a median survival of 446 days (range 41 to 2,308-k). Thirty-three patients who achieved a complete remission (Group Ila) had elevated transaminase levels secondary to presumed viral hepatitis. The other nine patients (Group Ilb) included seven who had no significant elevation in transaminase levels during the study period and two who were believed to have drug-related hepatotoxicity secondary to treatment with phenothiazines and estrogen plus cytarabine, respectively. The patient characteristics of Groups Ila and Ilb were similar (Table I). The majority of patients required only one course of induction chemotherapy to attain a bone

416

September

1983

The American

Journal

of Medicine

marrow remission, and 42 percent of both groups received levamisole as immunotherapy. The majority of blood cell component exposure in both groups was in the form of platelet concentrates. Type B hepatitis: Type B hepatitis developed in four patients (Table II). Three of these patients had hepatitis B surface antigen detected, while the fourth patient was diagnosed by seroconversion. Hepatitis B e antigen was found in all three hepatitis B surface antigen-positive patients; one of these three became a chronic hepatitis B surface antigen carrier. Antibody to hepatitis B surface antigen did not develop in this patient despite the presence of high-titer antibody to hepatitis B core an-

Volume

75

HEPATITIS AND ACUTE LEUKEMIA--WADE

tigen. The surface antigen was serotyped as ayw for all three patients in whom it was detected. The median maximum incubation period was 148 days, with a range of 102 to 170 days (Figure 1). The median duration of abnormal results of liver function tests was 142 days, with a median peak transaminase value of 142 mu/ml and a median bilirubin value of 1.4 mg/dl. All four patients were noted to be symptomatic with easy fatigability, nausea, and anorexia; hepatomegaly by physical examination was found in three of the four patients, but none of the patients had a prolonged prothrombin time, hypoalbuminemia, or ascites. Subsequent maintenance chemotherapy was delayed for a period of three months in two patients, and five and six months in the other two. New episodes of hepatitis B surface antigen-positive hepatitis developed later in the course in three other patients who had episodes of non-A/non-B hepatitis during initial remission induction and a fourth patient who was hepatitis-free during induction. Three had typeable surface antigen and were found to have the ayw serotype. All four of these patients were hepatitis B e antigen-positive; two became chronic hepatitis B surface antigen carriers. There were no active cases of Type A hepatitis: type A hepatitis noted, although 30 of the 42 patients had pre-existing antibody to hepatitis A virus on admission. Passive transfer of hepatitis A virus IgG antibody occurred in three additional patients. Cytomegalovirus, Epstein-Barr virus, herpes simplex, and Toxoplasma gondii hepatitis: Serologic evidence of hepatitis caused by these organisms de-

veloped in none of the 42 patients. Non-A/non-B hepatitis: Non-A/non-B hepatitis developed in 29 patients (Table II). Testing for hepatitis B surface antigen and IgM antibody to hepatitis B core antigen gave consistently negative results, while passive transfer of antibody to hepatitis B surface antigen and/or antibody to hepatitis B core antigen occurred in 15 patients, and two patients had antibody to both surface and core antigen prior to initiation of induction therapy. Type A hepatitis antibody was common, with 27 patients having pre-existing antibody to hepatitis A virus. Antibody to hepatitis A virus developed in the remaining two patients, but IgM determinations were negative. Antibody to hepatitis A virus was therefore believed to have been passively transferred. The median maximum incubation period was 77 days, with a range of 19 to 143 days (Figure 1). The median duration of abnormal transaminase values was 114 days, with a range of 10 to 518 days. The median peak transaminase value was 393 mu/ml (range 58 to 2,200), and the median peak bilirubin level was 4.0 mg/dl (range 0.7 to 19.1). Eighteen of the 29 patients were icteric, and 20 were symptomatic with nausea,

ET AL

200 190 180;;

170-

.

160-

.

0” P 150 0

14Q-

P E

130-

0.

---MEDIAN Protocol # 7405 0 Protocol # 7002

l

______ . .

0

.

80 8

a

120-

g IIOF 2 1002 gog r 1 z j;

80

2

50- 40

0

.. :

l

8

-A-

0

7060-

30 -

+ __*_-

‘0* l

k

.

?I0

:

20 -

0 0

IOI4 patients 1

(2 patients)

(29Patlents~

HEPATITIS E

Figure 1. titis.

Maximum

(30Patlents1

NON A/ NON B HEPATITIS

incubation

periods

for viral hepa-

anorexia, and easy fatigability, but only 13 were noted by physical examination to have hepatomegaly. None of the patients had prolonged prothrombin time, hypoalbuminemia, or ascites. Liver biopsies were performed in 10 of the 29 patients and consistently revealed hepatocellular swelling and disarray with a periportal mononuclear inflammatory infiltrate, but no fibrosis. Persistent hepatitis occurred in nine patients. Maintenance chemotherapy was delayed from six weeks to 26 months in 19 of the 29 patients. As previously described, hepatitis B surface antigen-positive hepatitis subsequently developed in three patients who experienced an episode of non-A/non-B hepatitis. Group Ill. Seventy-seven patients on UMCC protocol 7802 received induction chemotherapy. Fifty-one of these patients (66 percent) achieved a complete remission, with a median remission duration of 321 days (range 36 to 1,169-I) and a median survival of 33 1 days (range 70 to 1,230-l-). Thirty-two patients who achieved a complete remission (Group Illa) had elevated transaminase levels secondary to documented or presumed viral hepatitis. The other 19 patients (Group Illb) had no significant rise in transaminase levels during the study period. The patient characteristics (Table I) for Groups llla and lllb were similar, and did not differ significantly from patients in Group II. Once again, for these patients, the major blood product exposure was in the form of platelet concentrates.

September 1993

The American

Journal

of Medicine

Volume 75

417

HEPAT!TlS AND ACCTE !_C’JK?v’!A-WAD’

TABLE III -.___-_.-

5’

A’_

Relationship ‘of Viral Hepatitis to Duration of Hematologic Remission and Survival -_--__.--Numbelr of Patients

Protocol 7405 Group iia. Hepatitis (total) Type B Non-A/non-B Group lib. No hepatitis Protocol 7002 Group iia. Hepatitis (total) Type B Non-A/non-B Group iiib. No hepatitis

Daysto CompleteRemission’

33 4 29 9

40 47 38 35

(22-159) (29-59) (22-159) (24-58)

32 2 30 19

36 (27-126) 39; 126 36 (22-l 26) 35 (24-89)

Remission Duration+

247 218 258 125

@g-2,065+) (113-786) (39-2,065+) (6+-183)

330 (57-1,169+) 251; 640 338+ (58-1,169+) 312 (36-713)

Postrelapse Survival’

244 117 260 198

(74-962) (106-897) (74-962) (139-486)

169 (93-609) 246; 143+ 169 (93-609) 134 (34-351+)

Total Survival’

672 672 697 372

(119-2,1X+) (233-1,010) (119-2,115+) (41-617)

836 (170-1,230+) 497; 783-k 836 (170-1,230+) 495 (70-1,187+)

* Median days (range). + Patient died in complete hematoiogic remission.

Type B hepatitis developed in two of the 32 patients with hepatitis. Hepatitis 6 surface antigen was detected in both, although serotyping was not performed (Table II). The blood product exposure for these two patients was similar to Group III as a whole, and the hepatitis characteristics were similar to those cases of type B hepatitis observed in Group II patients. The median (mean) maximum incubation period for these two cases was 94 days, with a range of 71 to 119 (Figure 1). Neither of these two patients became chronic hepatitis B surface antigen carriers. A new episode of hepatitis that was hepatitis B surface antigen-positive developed later in the course in one other patient who, during the initial remission induction, had an episode of non-A/non-B hepatitis. Presumed non-A/non-B hepatitis developed in 30 patients, although the extensive serologic evaluation for other causes of infectious hepatitis was not carried out in Group Ill patients as it was for Group II patients (Table II). The mean number of transfused blood products for these patients was 86, somewhat but not sig-

I

nificantly higher than for the Group II patients with non-A/non-B hepatitis. The median maximum incubation period was 67 days (range 10 to 133) (Figure 1). The median peak transaminase value was 346 mu/ml (range 77 to 2,620), and the median peak bilirubin level was 1.6 mg/dl (range 0.4 to 20). The median duration of abnormal transaminase levels was 78 days (range 12 to 740), with persistent hepatitis present in seven patients. Liver biopsies were performed in seven of the 30 patients, and showed changes (hepatocellular swelling and periportal mononuclear infiltrate without fibrosis) consistent with viral hepatitis. The hepatitis characteristics in Group II and Group III patients were similar. Duration of Leukemia Remission. The patients in whom viral hepatitis developed (Groups la, Ila, and Illa) had a median time to complete remission that did not differ from that of the patients without hepatitis (Groups lb, Ilb, and Illb), nor did the time to remission for the patients with type B hepatitis vary significantly from those with non-A/non-B hepatitis (Table III). However,

L DAYS IN REMISSION

Duration of complete remisFigure 2. sion in patients with acute nonlymphocytic leukemia.

418

September 1983

The American Journal of Medicine

Volume 75

HEPATITIS AND ACUTE LEUKEMIA-WADE

the median complete remission duration in Group II patients was significantly longer for those with hepatitis (247 days) than for those without hepatitis (125 days) (p = 0.03) (Table Ill and Figure 2). The specific type of viral hepatitis or administration of levamisole was not important with regard to remission duration. There are presently three patients from Group II who remain in complete remission at 1,763-l-, 1,900-l-, and 2,258-I days. All three patients experienced non-A/non-B hepatitis during the initial induction period. In the patient with a remission duration of 1,900-l- days, type B hepatitis subsequently developed after the evaluation period; this patient has remained a chronic surface antigen carrier. The apparent advantage that the development of hepatitis had upon the complete remission duration for Group II patients was not found in patients from Group I or Group Ill (Figure 2). The median complete remission duration for Group III patients with hepatitis was 330 days (range 57 to 1,169-I) (Table Ill), which was similar to the remission duration for those patients without hepatitis (312 days [range 36 to 7131) (Figure 2). In an attempt to make Group II and Group Ill patients more comparable, an analysis of the development of hepatitis and remission duration was performed on only those patients in Group Ill who received chemotherapy alone as maintenance therapy. As seen in Figure 3, the development of hepatitis had no beneficial effect on remission duration. Survival afler Relapse of Leukemia. Survival following relapse of acute leukemia (Table Ill) was not significantly different between patients with hepatitis (Groups Ila and Illa) and the patients without hepatitis (Groups Ilb and Illb) (Table III). Overall Survival. The median duration of survival was significantly longer (p = 0.01) for Group II patients with hepatitis (672 days) than for those patients without hepatitis (372 days) (Figure 4). Once again, the type of viral hepatitis was not a factor (type B, 672 days; nonI

I

s

( Protocol

g

7802

ET AL

1

4t

i,;

,

,

00

200

Days

, 300

400

in REMISSION

Figure 3. Duration of complete remission for Group 111 patients randomized to chemotherapy alone for remission maintenance.

A/non-B, 697 days) (Table Ill). There are presently five patients who remain alive, with survival durations of 1,792-l-, 1,932+, 1,960-I, 2,090+, and 2,308+ days. All five of these patients contracted non-A/non-B hepatitis, although two later in their course have contracted type B hepatitis and have remained chronic hepatitis B surface antigen carriers. Group III patients in whom hepatitis developed experienced no prolongation of survival (Figure 4), with a median duration of 836 days (range 170 to 1,230-t) (Table Ill) for patients who contracted hepatitis compared with 495 days (range 70 to 1,187+) for patients without hepatitis. Degree of Hepatic Injury. Analysis of the role of hepatic injury was evaluated by comparing the height of the peak transaminase level, the duration of abnormal transaminase levels, and peak bilirubin values with the

Group II (Protocol

7405)

I

__

Days of Survival

Figure 4. Duration of survival in patients with acute nonlymphocytic leukemia.

September 1983

The American Journal of Medicine

Volume 75

419

UE=‘AT’T’S AND ACC’TE ‘LEC’YEM’A-WADE CT A’_

cluraition of hematologic: remission, postrelapse survival, and total survival. INo correlation could be demonstrated for these paramete~rs. COMMENTS Patients with acute nonlymphocytic leukemia who undergo induction chemotherapy frequently experience elevations in transaminase levels. The cause of these hepatic abnormalities is often unclear, and their presence may be responsible for delays in further administration of antineoplastic therapy. Recent reports of longer survival in patients with acute nonlymphocytic leukemia who had elevations in transaminase levels [8,21] led us to review our experience with three homogeneous groups of patients with acute leukemia. Hepatitis occurred in 73 percent of these patients who received many transfusions, with only 6 percent of the hepatitis being identified as type B. The 6 percent incidence of type B hepatitis is consistent with recent prospective studies of post-transfusion hepatitis [l3,6,7,22]. Type B hepatitis in these patients had a median maximum incubation period consistent with a previous report [23]. Although these patients were symptomatic, the mean peak transaminase value was minimal, and only one of the patients was clinically icteric. The patients in whom typeable hepatitis B surface antigen-positive hepatitis developed (Group II) all had the same surface antigen serotype, ayw. This serotype is infrequent and its exclusive occurrence is suggestive of a common source, presumably a donor with chronic antigenemia who was used frequently during UMCC protocol 7405 [ 231. Hoofnagle et al [ 181 have proposed that donor blood that is hepatitis B surface antigennegative may be infectious if it possesses high-titer hepatitis B core antibody with negative surface antibody. Krugman et al [24] have disagreed, and suggested that antibody to hepatitis B core antigen is not an indicator of infectivity but rather immunity. It is of interest that three of the six patients (Group II) who acquired hepatitis B surface antigen had high levels of antibody to hepatitis B core antigen with negative antibody to hepatitis B surface antigen titers four to six months prior to the detection of hepatitis B surface antigen. This is consistent with the hypothesis of Hoofnagle et al [ 181. Ninety percent of the post-tranfusion hepatitis in this study was non-A/non-B, which is consistent with other reports [ 16,17,25-271. Although this review, like all others, lacks a definitive means for serologic confirmation, we nevertheless believe that these cases represent non-A/non-B hepatitis. The median maximum incubation period of 74 days is consistent with other reports [ 16,17,25,27,28]. The majority of our patients were symptomatic, with greater than 50 percent experiencing icterus, in contrast to the approximately 25

420

September 1983

The American Journal of Medicine

percent usually reported [ 1,2,8,29]. Persistent hepatitis developed in 27 percent (nine of 29 patients in Group II and seven of 30 patients in Group III, similar to the report by Knodell et al [30]. In the 17 patients who underwent liver biopsy, bridging necrosis and fibrosis were uniformly absent, which is consistent with other reports

[311. Like the experience of other investigators, agents such as cytomegalovirus, Epstein-Barr virus, herpes simplex virus, or Toxoplasma gondii were not found to be the cause of non-A/non-B hepatitis [ 2,301. No cases of acute type A hepatitis were detected, although within the Group II patient population 82 percent had antibody at the time of initial admission, which is consistent with previous seroepidemiologic studies that showed a high proportion of middle-aged adults with type A hepatitis antibody [ 321. This 70 percent incidence of hepatitis is high but not significantly different from the 64 percent reported by Barton and Conrad [8] in a similar patient population. As in the study by Barton and Conrad [8], no difference in the number of units of blood components transfused was observed between the group with hepatitis and the group without hepatitis. The majority of the blood product transfusions were supplied as platelet concentrates from an intensively screened (i.e., determination of hepatitis B surface antigen by radioimmunoassay, measurement of serum bilirubin and serum glutamic oxalacetic transaminase levels, plus history and physical examination) commercial donor pool. Despite the high frequency of hepatitis, the common delay in the administration of subsequent chemotherapy, and the potential hazard of transmission, we have seen rather minimal hepatitis-associated morbidity, but unlike Barton and Conrad [8] and Foon et al [21], we have not noted a consistent, significant prolongation of survival or complete remission duration. The reason for this disparity remains unclear. Group II patients in whom hepatitis developed had a substantial increase in the median duration of complete remission and an increase in survival. It could be questioned whether this difference is simply a phenomenon of the patients who lived longer being the ones at greater risk for the development of hepatitis rather than the hypothesis that the development of hepatitis confers an advantage with regard to remission duration. Hence, statistical analysis requires a correction for this time-dependent covariable. This was a consideration not addressed by Barton and Conrad [8] and Foon et al [21] in their studies. When this correction is made for Group II patients, the apparent advantage of the development of hepatitis on remission duration and hence on ultimate survival is confirmed. When Group III patients are evaluated utilizing these analysis techniques,

Volume 75

HEPATITIS AND ACUTE LEUKEMIA--WADE

no benefit from the development of hepatitis exists. The major difference in these two antileukemic protocols is the maintenance therapy, and is substantiated by the similar remission induction rates (63 percent versus 66 percent) but the longer the complete remission duration for Group III patients (321 days versus 183 days). Perhaps any benefit provided by the development of hepatitis becomes hidden in the benefits of more intensive maintenance chemotherapy combined with splenectomy or neuraminidase-treated allogenic myeloblasts. If the development of hepatitis truly prolongs duration of complete remission, one would expect to see this advantage in the subset of Group III randomized to chemotherapy alone. The total number of patients in this

ET AL

group is small, but at present no advantage exists for those patients in whom hepatitis develops. In conclusion, we report three groups of prospectively treated patients with acute nonlymphocytic leukemia who underwent intensive transfusion and who showed a high incidence of non-A/non-B hepatitis. These patients, although frequently symptomatic and icteric, experienced minimal morbidity, but a consistent prolongation of complete remission and survival could not be demonstrated. The development of serologic methods to detect non-A/non-B hepatitis in blood product donors hopefully will decrease the occurrence of this illness in patients requiring intensive blood product support.

REFERENCES

2.

3.

4.

5.

6.

7.

8. 9.

10.

11.

12.

Aach RD, Lander JJ, Sherman LA, et al: Transfusion-transmitted viruses: interim analysis of hepatitis among transfused and nontransfused patients. in: Vyas GN, Cohen SN, Schmid R, eds. Viral hepatitis. Philadelphia: Franklin Institute Press, 1978; 383-396. Alter HJ, Purcell RH, Feinstone SM. Holland PV, Morrow AG: Non-A/non-B hepatitis: a review and interim report of an ongoing prospective study. In Ref 1; 359-381. Aach RD. Shields HM, Lander JJ, Perrillo RP, Zuckerman GR: Post-transfusion hepatitis leading to chronic hepatitis. Gastroenteroiogy 1978; 75: 732-741. Seeff LB, Wright EC, Zimmerman HJ, et al: Post-transfusion hepatitis, 1973-1975: a Veterans Administration cooperative study. In Ref 1; 371-382. Prince AM, Grady GF, Hazzi C, et al: Long-incubation posttransfusion hepatitis without serological evidence of exposure to hepatitis-B virus. 1974; li:241-246. Prince AM, Brotman B, Grady GF et al: Post-transfusion viral hepatitis caused by an agent or agents other than hepatitis B virus or hepatitis A virus: impact on efficacy of present screening methods. In: Greenwalt TJ, Jamieson GA, eds. Transmissible disease and blood transfusion. New York: Grune & Stratton, 1975; 129-140. Goldfield M, Black HC, Bill J, Srihongse S, Pizzuit W: The consequence of administering blood pretested for HB,Ag by third generation techniques: a progress report. Am J Med Sci 1975; 270: 335-342. Barton JC, Conrad ME: Beneficial effects of hepatitis in patients with acute myelogenous leukemia. Ann Intern Med 1979; 90: 188-190. Smyth AC, Schimpff SC, Wiernik PH: lnctience and chronicity of HB,Ag negative hepatitis in acute nonlymphocytic leukemia (ANLL). Clin Res 1975; 23: 343A. Wiernik PH, Schimpff SC, Schiffer CA, et al: Randomized clinical comparison of daunorubicin (NSC-82151) alone with a combination of daunorubicin, cytosine arabinoside (NSC-63878), 6-thioguanine (NSC-752) and pyrimethamine (NSC-3061) for the treatment of acute nonlymphocytic leukemia. Cancer Treat Rep 1976; 60: 41-53. Ling CM, Overby LR: Prevalence of hepatitis B virus antigen as revealed by direct radioimmune assay with 1251-antibody. J lmmunol 1972; 109: 834-841. Peterson MR, Barker LF. Schade IX: Detection of antibody to hepatitis-associated antigen in hemophilia patients and in voluntary blood donors. VOX Sang 1973; 24: 66-75.

13.

14.

15.

16.

17.

18.

19. 20.

21.

22. 23.

24.

25.

26. 27.

September 1983

Overby LR, Ling CM: Radioimmune assay for anti-core as evidence of exposure to hepatitis B virus. Rush-Presbyt-St. Luke’s Med Bull 1976; 15: 83-92. Purcell RH, Wong DC, Morilsugu Y, Deinstag JL, Routenberg JA, Boggs JD: A microhler solid phase radioimmunoassay for hepatitis A antigen and antibody. J lmmunol 1976; 116: 349-356. Bradley DW, Maynard JF, Hindman SH, et al: Serodiagnosis of viral hepatitis A-detection of acute phase immunoglobulin Mj anti-hepatitis A virus by radioimmunoassay. J Clin Microbiol 1977; 5: 521-530. Alter HJ, Holland PV, Purcell RH: The emerging pattern of post-transfusion hepatitis. Am J Med Sci 1975; 270: 329-334. Aach RD, Kahn RA: Post-transfusion hepatitis: current perspectives. Ann intern Med 1980; 92: 539-546. Hoofnagle JH. Seeff LB, Bales ZB, Zimmerman HJ, and the Veterans Administration Hepatitis Cooperative Study Group: Type B hepatitis after transfusion with blood containing antibody to hepatitis B core antigen. N Engl J Med 1978; 298: 1379-1383. Cox DR: Regression models and life tables. J R Stat Sot 1972; 34: 187-196. Mantel N, Byar DP: Evaluation of response-time data involving transient states. An illustration using heart transplant data. J Am Stat Assoc 1974; 69: 81-87. Foon KA, Yale C, Coldfelter K, Gale RP: Posttransfusion hepatitis in acute myelwnous leukemia. JAMA 1980; 244: 1806-1807. Wands JR, Walker JA, Davis TT, et al: Hepatitis B in an oncology unit. N Engl J Med 1974; 291: 1371-1374. Holland PV, Purcell RH, Smith H, Alter MI: Subtyping of hepatitis-associated antigen (HBAg): simplified technique with counterelectrophoresis. J lmmunol 1972; 109: 420425. Krugman S, Overby LR, Mushahwar IK, Ling CM, Frosner GG, Deinhardt F: Viral hepatitis, type B: studies on natural history and prevention re-examined. N Engl J Med 1979; 300: 101-106. Feinstone SM. Kapikian AZ, Purcell RH, Alter HJ, Holland PV: Transfusion-associated hepatitis not due to viral hepatitis type A or B. N Engl J Med 1975; 292: 767-770. Gocke DJ: Viral hepatitis. Postgrad Med 1975; 58: 137142. Tabor E, Gerety RJ: Non-A, non-B hepatitis: new findings and prospects for prevention. Transfusion 1979; 19: 669-

The American Journal of Medicine

Volume 75

421

HiPATITIS

AND AC’JT!! LE’JKEM’A-WADE

CT A’_

674.

28.

29.

30.

422

Hoofnegle JH, Gerety RJ, Tabor E, Feinstone SM, Barker LF, Purcell RH: Transmission of non-A, non-B hepatitis. Ann Intern Med 1977; 87: 14-20. Seeff LB, Wright EC, Zimmerman HJ, et al: Veterans Administration cooperative study of post-transfusion hepatitis, 1969-1974: incidence .and characteristics of hepatitis and responsible risk factors. Ann J Med Sci 1975; 270: 355362. Knodell RG. Conrad ME, Ginsberg AL, Bell CJ, Flannery EP:

September 1983

The American Journal of Medicine

31.

32.

Volume 75

Efficacy of prophylactic gamma-globulin in preventing non-A, non-B post-transfusion hepatitis. Lancet 1976; I: 557-561. Berman M, Alter I-N, lshak KG, Purcell RH, Jones EA: The chronic sequelae of non-A, non-B hepatitis. Ann intern Med 1979; 91: l-6. Szmuness W, Dienstag JL, Purcell RI-l, Harley EJ, Stevens CE, Wong DC: Distribution of antibody of hepatitis A antigen in urban adult populations. N Engl J Med 1976; 295: 755-759.