1. Hepatitis B virus reactivation during immunosuppressive drug therapy
Robert G Gish MD
Professor Consultant
Stanford University

2. HBV morbidity & mortality
> 2 billion have been infected
4 million acute cases per year
1 million deaths per year
240 million chronic carriers
25% of carriers die from chronic hepatitis, cirrhosis, or liver cancer
Nearly 75% of chronic carriers are Asian
Second most important carcinogen behind tobacco
Causes 60% to 80% of all primary liver cancer
HBV is 100 times more contagious than HIV

3. Definitions
Virologic increase of 1 log IU/ml or de novo appearance of HBV DNA when previously non detectable
Proposed: 2 log increase or de novo or reappearance of HBV DNA to a level of at least 100 IU/mL (AASLD Emerging Trends Conference on HBV Reactivation, March, 2013)
In absence of HBV DNA measurements, reappearance of HBeAg or HBsAg is reasonable evidence
Increase of 3 fold or greater in ALT levels if BSL levels normal or 2 fold or greater increase over BSL if initially abnormal.

4. HBV reactivation Definition Outcomes of interest
De novo detection of HBV DNA
> 10 fold increase in HBV DNA level compared to baseline
Hepatitis flare: 2 – 3 fold elevation of ALT; > 100 U/L
Seroreversion (e.g., from negative HBsAg to positive)
Outcomes of interest
HBV reactivation
Hepatitis flare
HBV related hepatic failure
HBV related death
Chemotherapy disruption

5. Who will reactivate? + + + + + HBsAg Anti-HBc total Anti-HBs IgM
What it means A _ _
Never exposed B _ +
Vaccinated C _ +
Exposure D + _
Acute HBV E + _
Chronic HBV F _ +
Exposure
“Isolated core”

6. Burden of HBV Infection in the US
> 14 million (~ 1 out of 20) persons in the US have been infected[1]
An estimated ~ 2 million are chronically infected in the US[2]
73,000
2,000,000
New Acute Infections per Year[3]
Chronic Infections[2]
HCC, hepatocellular carcinoma.
The first slide in this group evaluates the burden of HBV infection in the United States. This is important because more than 14 million people have been exposed to HBV, having at least 1 positive hepatitis B test. Currently, it is estimated that 2 million individuals in the United States are HBsAg positive.
Considerable data indicate that the potential burden associated with HBV infection in the United States is currently underestimated. Although acute infections do continue to decline, I believe that the number of deaths per year is underestimated because hepatitis B is not identified as the cause on death certificates. This is also true with liver cancer, such that a patient with hepatocellular carcinoma (HCC) may die of liver failure but the presence of HCC may not be identified before death—that is, not until autopsy or another later time. Consequently, HCC may not be listed in the medical information of the patient, leading to lower reported rates. The number of liver transplantations is low and continues to decline. Perhaps the reason the number of liver transplantations is not higher is because patients are diagnosed with liver cancer late, when they are no longer candidates for transplantation. Of course, all liver cancer patients could potentially be thought of as transplantation candidates. But with improved screening and surveillance, it is likely that more patients will receive a transplantation earlier because of liver cancer.
5000
3100
230
Deaths per Year[3]
Diagnosed HCC per Year[4]
Liver Transplants per Year[5]
1. CDC. Available at: Accessed May 20, Cohen C, et al. J Viral Hepat. 2008;15: CDC. HBV disease burden El-Serag HB, et al. Arch Intern Med. 2000;160: UNOS/OPTN. 6

7. Acute HBV infection
Universal vaccination of infants recommende d in 1991
80% decline in incidence

8. Natural History of HBV Reactivation During Chemotherapy
ChemoRx and/or
steroids
HBV DNA
IMMUNE
SUPPRESSION
Recovery of neutropenia or steroid withdrawal
Death
Acute liver failure
Chronic hepatitis
Cirrhosis
RECOVER
Acute
hepatitis
ALT
IMMUNE REBOUND 4 8 12 16 20 24 28 32 36 52
100
Weeks after Exposure

9. Risk Factors for HBV Reactivation
Patient and HBV status
Male, young age
Level of HBV-DNA >20,000 IU/ml
HBeAg positive
Active liver disease
Anti-HBc
Level of immunosuppression
Corticosteroids
Chemotherapy regimens : Vinca-alkaloid, anthracycline
Rituximab, infliximab
Immune reconstitutional periods
Bone marrow transplantation
TACE
Mindikoglu AL, Regev A, Schiff ER. Clin Gastroenterol Hepatol 2006;4:
Manzano-Alonso ML, Castellano-Tortajada G. World J Gastroenterol 2011;17:

10. Anti-HBc and reactivation
FDA Boxed Warning Sept
109 fatal cases of HBV acute liver injury
106 with rituximab, 3 with ofatumumab
69% cases were HBsAg negative prior to initiation of immune suppression
Reactivation 63 days to 12 months after last dose
Recommendation:
Screen ALL patients before treatment with HBsAg, anti-HBc
In patients with anti-HBc+ and HBsAg(-) consider HBV DNA NAT
Refer to specialist
Monitor for “several months” after treatment

11. Content Conflicting practice recommendations
Review using GRADE criteria
Risk of HBV reactivation (HBVr) for immunosuppressive agents
Prevention and treatment of HBVr

12. Conflicting practice recommendations
American Society of Clinical Oncology (2010) has published a provisional clinical opinion which states that there is insufficient evidence to recommend routine screening in patients undergoing immunosuppressive therapy, updated 2015 that was more confusing
Screening recommended by CDC (2008), AASLD (2009), APASL (2012), and EASL (2012)
No guidelines from American Academy of Dermatology and American College of Rheumatology
Surveys have shown that routine HBV screening is done by
70% of rheumatologists
40% of dermatologists
20-40% of oncologists
GI? Other?

13. Methods AGA technical review and practice guideline
Followed best practices outlined by the IOM (2011)
“Clinical Practice Guidelines We Can Trust”
Establishing transparency
Management of conflict of interest
Evidence based on systematic reviews
Method for rating quality of evidence (GRADE)
Method for rating strength of recommendations (GRADE)
Articulation of recommendations
External review

14. Literature search

15. Analytical approach Pooled effects from RCTs (random effects model)
Initially tested hypothesis that relative treatment effects are usually stable across different baseline risk
Little or no heterogeneity observed
Applied pooled relative risks to typical baseline risk from different populations (those from available RCTs but also from other settings) to arrive to representative risk differences to inform recommendations

16. High risk for HBVr (>10%)
B-cell depleting agents (rituximab, ofatumumab)
HBsAg +
core +
30 – 60%
HBsAg -
core +
17%
Confidence in estimate A B
Anthracycline derivatives (e.g.,doxorubin / epirubicin)
15 – 30%
Corticosteroids for > 4 wks (> 10 mg prednisone)
> 10%
Categories of confidence in the estimate:
(A) High confidence that the estimate lies within group risk boundaries
(B) Moderate confidence that the estimate lies within group risk boundaries
(C) Little or no confidence that the estimate lies within group risk boundaries

17. Rituximab-Associated HBV Reactivation in Lymphoproliferative Disorders
Meta-analysis and review of FDA safety profiles
Case reports (n=27)
Case series reports (n=156)
Onset post last rituximab dose
Median: 3 months (range: months)
>6 months: 29%
Reactivation in anti-HBcpositive patients receiving rituximab versus no rituximab
Odds ratio: 5.73 (P=0.0009)
HBV Reactivation Risk:
Rituximab-Treated Lymphoma Patients
Hui
(n=233)
Tsutsumi
(n=47)
Targhetta
(n=319)
Yeo
(n=50)
Fukushima
(n=48)
Overall
(n=697)
12.44
5.24
3.38
9.39
Slide: Rituximab-Associated HBV Reactivation in Lymphoproliferative Disorders
Evens and colleagues conducted a meta-analysis of all published cases of HBV reactivation and those reported to Food and Drug Administration (FDA) Adverse Event Reporting System (AERS) MedWatch database. A total of 183 unique cases were identified.1
HBV reactivation from rituximab typically occurred 3 months after the last dose of rituximab, although 29% of the cases occurred more than 6 months post therapy.1
Overall odds ratio: 5.73 (95% CI ; P=0.0009).
Odds ratio in lymphoma patients: 5.64 (95% CI ).
Reference
Evens AM, Jovanovic BD, Su YC, et al. Rituximab-associated hepatitis B virus (HBV) reactivation in lymphoproliferative diseases: meta-analysis and examination of FDA safety reports. Ann Oncol.2010;November 29. [Epub ahead of print].
1.60
5.64
( )
Odds Ratio
Evens AM, et al. Ann Oncol.22: , 2011

18. Rituximab risk for HBVr
Pooled baseline risk estimate without prophylaxis of HBVr with rituximab in patients who have recovered from hepatitis B infection (HBsAg-negative, anti-HBc-positive).
Perrillo Gastro 2015

19. Baseline Characteristics
HBV Reactivation in Patients Treated With Anti-Tumor Necrosis Factor Agents
Retrospective study ( )
Kaiser Permanente Northern California
Patients on anti-TNF (n=8887)
Followed until death, liver transplant, or end of membership
Electronic medical review
All cases of grade 3/4 hepatotoxicity
AST/ALT >5x ULN or 5x baseline
HBV reactivation
1 log increase in HBV DNA from baseline
HBV DNA >2000 IU/mL (if no baseline)
HBV DNA positive when previously negative
HBsAg positive when previously negative
Baseline Characteristics
Patients
(n=8887)
Age (years) 49
Male (%) 38
Time on anti-TNF (months) 41
Treated by (%)
Dermatology
Rheumatology
Gastroenterology
More than 1 24 53 13 10
Steroids + immunosuppressants (%)
Anti-TNF agent (%)
Infliximab
Adalimumab
Etanercept 12 33 32
Slide: HBV Reactivation in Patients Treated With Anti-Tumor Necrosis Factor Agents
Pauly and colleagues conducted a retrospective study ( ) to determine rate of HBV testing at baseline and the incidence of HBV reactivation and hepatotoxicity in a large cohort of patients treated with anti-tumor necrosis factor (anti-TNF) (n=8887).1
Kaiser Permanente Northern California.
Patients followed until death, liver transplant, or end of membership.
Electronic medical review to identify all cases of grade 3/4 hepatotoxicity (defined by AST/ALT >5x ULN or 5x baseline) and HBV reactivation (defined by 1 log increase in HBV DNA from baseline, HBV DNA >2000 IU/mL [if no baseline], HBV DNA positive when previously negative, or HBsAg positive when previously negative).
Reference
Pauly MP, Tucker L-Y, Szpakowski J-L, et al. Reactivation of hepatitis B (HBV) in patients treated with anti-tumor necrosis factor (anti-TNF) agents. Hepatology. 2014;60(suppl 1):232A-233A. Abstract 72.
Pauly MP, et al. Hepatology.2014;60(suppl 1):232A-233A Abstract 72.

20. HBV Testing Status and Reactivation in Patients Treated With Anti-Tumor Necrosis Factor Agents
HBV testing at baseline increased from 37% (2001) to 72% (2010)
HBV testing status of overall cohort at baseline
Tested/never tested/others: 52%/27%/21%
HBV reactivation (n=9)
HBsAg positive (n=7, none received prophylactic antiviral therapy)
Silent reactivation (n=5)
Grade 3/4 hepatotoxicity (n=2)
HBsAg negative, HBcAb negative (n=1; silent reactivation)
Other (n=1, not tested at baseline, grade 3/4 hepatoxicity)
Clinical approach to reactivation
Discontinued anti-TNF therapy (n=3)
HBV antiviral therapy, all responded (n=6)
No hospitalizations, liver failure, or death
HBV Status at Baseline
Percent of
Total cohort (n=8887)
Tested for HBV (n=4621)
95.7%
Patients (%)
49.7%
Slide: HBV Testing Status and Reactivation in Patients Treated With Anti-Tumor Necrosis Factor Agents
From the period of 2001 to 2010, HBV testing at baseline increased from 37% to 72%. HBV testing status of overall cohort at baseline was:1
Tested/never tested/others: 52%/27%/21%.
A total of 9 cases of HBV reactivation were identified.1
HBsAg positive (n=7, none received prophylactic antiviral therapy).
Silent reactivation (n=5).
Grade 3/4 hepatotoxicity (n=2).
HBsAg negative, HBcAb negative (n=1; silent reactivation).
Other (n=1, not tested at baseline, grade 3/4 hepatoxicity).
The clinical approach to patients with HBV reactivation included discontinuation of anti-TNF therapy in 3 cases and initiation of HBV antiviral therapy in 6 cases (all responded). There were no hospitalizations, liver failure, or deaths among the HBV reactivation cases in this cohort.1
Reference
Pauly MP, Tucker L-Y, Szpakowski J-L, et al. Reactivation of hepatitis B (HBV) in patients treated with anti-tumor necrosis factor (anti-TNF) agents. Hepatology. 2014;60(suppl 1):232A-233A. Abstract 72.
3.9%
0.5%
2.0%
0.3%
HBsAg
Positive
HBcAb Positive
HBsAg Negative
HBsAg Negative
HBcAb Negative
Pauly MP, et al. Hepatology.2014;60(suppl 1):232A-233A Abstract 72.

21. HBV Reactivation in Patients Treated With Anti-Tumor Necrosis Factor Agents
Grade 3/4 hepatotoxicity (2.7%; 243/8887)
HBV (n=3; no hospitalizations or deaths)
Indeterminate (n=100)
Other (n=140)
Rate of HBV reactivation and grade 3/4 hepatoxicity was low, but both were higher among those who were HBsAg positive at baseline
HBV reactivation in this cohort did not result in serious complications
Conclusions
HBsAg positive patients should be evaluated for prophylactic antiviral therapy or be monitored closely for HBV reactivation
HBsAg negative/HBcAb positive patients should be monitored and HBV testing performed if grade 3/4 hepatotoxicity develops
Further studies are needed to identify best HBV screening strategies among patients taking anti-TNF therapy
Slide: HBV Reactivation in Patients Treated With Anti-Tumor Necrosis Factor Agents
Grade 3/4 hepatotoxicity developed in 2.7% of patients in the cohort and among these 3 were caused by HBV, 100 were of indeterminate causes, and 140 were of other causes.1
The rate of HBV reactivation and grade 3/4 hepatoxicity was low, but the rates of both outcomes were higher among those who were HBsAg positive at baseline. HBV reactivation in this cohort did not result in serious complications.1
Based on these results, the investigators recommended that HBsAg positive patients should be evaluated for prophylactic antiviral therapy or be monitored closely for HBV reactivation. For HBsAg negative/HBcAb positive patients, they should be monitored and HBV testing performed if grade 3/4 hepatotoxicity develops. Further studies are needed to identify best HBV screening strategies among patients taking anti-TNF therapy.1
Reference
Pauly MP, Tucker L-Y, Szpakowski J-L, et al. Reactivation of hepatitis B (HBV) in patients treated with anti-tumor necrosis factor (anti-TNF) agents. Hepatology. 2014;60(suppl 1):232A-233A. Abstract 72.
Pauly MP, et al. Hepatology.2014;60(suppl 1):232A-233A Abstract 72.

22. Prophylaxis of HBV Reactivation With Tenofovir DF
Ongoing, prospective, open-label study (n=69)
HBeAg-positive and -negative patients
HBV DNA undetectable before starting rituximab for hematologic malignancies
HBV reactivation
HBV DNA elevation >1 log10 IU/mL above baseline and/or HBsAg reappearance
Preliminary analysis of first 12 months in 30 of 69 patients enrolled
Prophylaxis arms (18 months)
Tenofovir DF
Observation
Baseline Characteristics
Tenofovir DF
(n=18)
Observation
(n=12)
Age (years) 66 73
Male (%) 61 58
Weight (kg)
71.7
75.4
BMI (kg/m2)
26.1
28.5
Anti-HBs positive (%)
55.6
50.0
Non-Hodgkin lymphoma (%)
77.8
83.3
Chronic lymphatic leukemia (%)
16.7
Slide: PREBLIN Study: Prophylaxis of HBV Reactivation With Tenofovir DF
Buti and colleagues are conducting an ongoing, prospective, open-label study (n=69) to assess the efficacy and safety of tenofovir DF in the prevention of HBV reactivation in anti-HBc positive patients with hematologic malignancies treated with rituximab.1
HBeAg-positive and -negative patients.
HBV DNA undetectable before starting rituximab for hematologic malignancies.
HBV reactivation was defined as an HBV DNA elevation >1 log10 IU/mL above baseline and/or HBsAg reappearance.
This preliminary analysis covers the first 12 months in 30 of 69 patients who were enrolled in the study.1
- Prophylaxis arms (18 months) were either tenofovir DF or observation.
Reference
Buti M, Manzano ML, Morillas RM, et al. Effect of hepatitis B virus reactivation on the recurrence of HBV-related hepatocellular carcinoma after curative resection in patients with low viral load. Hepatology. 2014;60(suppl 1):997A. Abstract 1661.
Buti M, et al. Hepatology.2014;60(suppl 1):997A Abstract 1661.

23. Preliminary Results of Tenofovir DF as Prophylaxis of HBV Reactivation
Preliminary analysis showed tenofovir DF prophylaxis prevented HBV reactivation
No statistically significant difference in liver and renal function between the 2 arms
HBV reactivation in the observation arm (n=2)
Elderly man (anti-HBs positive) and women (anti-HBs negative)
HBV DNA elevation >1 log10 IU/mL from baseline at 4 month visit
ALT <40 U/L
Rescue therapy
Tenofovir DF, both patients HBV DNA undetectable at 6-month visit
HBV Reactivation
(12-Month Preliminary Analysis)
16.7%
P=0.152
Patients (%)
Slide: PREBLIN Study: Preliminary Results of Tenofovir DF as Prophylaxis of HBV Reactivation
Preliminary analysis showed tenofovir DF prophylaxis prevented HBV reactivation in 18 of 18 patients. There was no statistically significant difference in liver and renal function between the 2 arms.1
Two HBV reactivation cases occurred in the observation arm. One patient was an elderly man (anti-HBs positive) and the other an elderly women (anti-HBs negative), both of whom had HBV DNA elevation >1 log10 IU/mL from baseline at the 4 month visit and an ALT <40 U/L. Both patients initiated rescue therapy with tenofovir DF and achieved undetectable HBV DNA levels at the 6-month visit.1
Reference
Buti M, Manzano ML, Morillas RM, et al. Effect of hepatitis B virus reactivation on the recurrence of HBV-related hepatocellular carcinoma after curative resection in patients with low viral load. Hepatology. 2014;60(suppl 1):997A. Abstract 1661. 0%
Tenofovir DF
(n=18)
Observation
(n=12)
Buti M, et al. Hepatology.2014;60(suppl 1):997A Abstract 1661.

24. Moderate risk (1% - 10%) HBsAg + core + HBsAg - core + 1% (C)
TNF-alpha inhibitors
(etanercept, adalimumab, certolizimab, infliximab)
HBsAg +
core +
1 – 10% (B)
HBsAg -
core +
1% (C)
1 – 10% (C)
(>10 mg)
Other cytokine & integrin inhib.
(abatacept, ustekinimab, natalizumab, vedolizumab)
1 – 10% (C)
Tyrosine kinase inhibitors
(imatinib, nilotinib)
1 – 10% (B)
TNF 3 mio on TNF, if HBsAgpos in gen pop is 0.3% and core pos is 3% = high number of reactivation possible
Corticosteroids for > 4 weeks
1 – 10% (B)
(low dose: <10 mg)
Anthracycline derivatives (e.g.,doxorubin / epirubicin)
1% (C)

25. Low risk group (<1%) HBsAg + core + HBsAg - core + <1% (A)
Traditional immunsupp. agents
(azathioprine, 6-mercaptopurin, methotrexate alone)
<1% (A)
<<1% (A)
Intra-articular corticosteroid
<1% (A)
<<1% (A)
Corticosteroids for ≤ 1 week
<1% (B)
<<1% (A)
Corticosteroids for > 4 weeks
<1% (B)
(low dose: <10 mg)
Categories of confidence in the estimate:
(A) High confidence that the estimate lies within group risk boundaries
(B) Moderate confidence that the estimate lies within group risk boundaries
(C) Little or no confidence that the estimate lies within group risk boundaries

26. Case 1
A 48 year-old lady who is HBsAg-negative and anti-HBc- positive is diagnosed with stage 3 non-Hodgkin’s lymphoma Further testing reveals that serum HBV DNA is negative. The oncologist recommends that she be placed on CHOP-R (rituximab) and it is planned that the patient will be kept on rituximab maintenance for six months after the last CHOP cycle. The Oncologist is concerned about a flare of HBV and seeks a GI consult.

27. Initiate anti-viral prophylaxis now and continue for a duration of 6 months from completion of CHOP-R therapy
Start monitoring HBV DNA monthly and intervene when there is a flare of HBV
 Initiate anti-viral prophylaxis now and continue for a duration of 12 months from completion of CHOP-R therapy
Use anti-viral prophylaxis indefinitely
No further intervention is not needed

28. Case 2
A 28-year-old Asian man suffers from frequent bouts of asthma. He is known to be HBsAg positive but has normal transaminases and in the past had negative HBV DNA.Hehas a family H/O hepatocellular carcinoma. He has frequent bouts of asthma and his PCP wishes to use corticosteroids for asthma exacerbation. When there is an exacerbation of asthma the PCP would like to place him on a 2-week course of prednisone (20 mg daily for one week followed by 10 mg for one week). The PCP would like your opinion regarding his HBV and whether there would be a concern.

29. use HBV anti-viral therapy for the duration of prednisone treatment
use HBV anti-viral therapy indefinitely
 obtain HBV DNA and if present, use HBV therapy for the duration of prednisone use
HBV therapy is not indicated
Obtain HBeAg and intervene with therapy if positive

30. Case 3 A 58 year old HBsAg-positive lady has disabling
plaque psoriasis and arthropathy
and normal serum aminotransferase
levels and has been taking
methotrexate 7.5 mg three times weekly for several years.
Recently, the plaques psoriasis has become resistant to this treatment
and the dermatologist advises that she be started on adalimumab.
You are asked to consult on this patient to provide further recommendations.

31. Initiate HBV prophylaxis to prevent a flare
As there was no flare on methotrexate, there was no concern with adalimumab
 monitor HBV DNA monthly and initiate HBV therapy if there is a flare
obtain HBeAg and initiate prophylaxis if it is positive 

32. Case 4
A 52 year-old Asian woman is diagnosed with lymph node positive adenocarcinoma of the breast and is advised to undergo surgery. She subsequently is advised to undergo chemotherapy when found to be HER-2 positive and at high risk of recurrence. The oncologist plans to use sequential anthracycline-taxane therapy. The patient is screened for HBV and found to be HBsAg (+) and anti-HBc (+) and is HBV DNA and HBe Ag negative.Liver profile is normal.

33. Check HBV DNA and initiate prophylaxis if elevated to above 20,000 IU/mL
 Initiate HBV prophylaxis and continue it for 6 months until after completion of therapy
Monitor HBV DNA monthly and initiate HBV therapy if it becomes positive
monitor liver profile monthly and have it trigger HBV therapy

34. Case 5
A 55 year old lady presents with left neck swelling. Chest X-Ray notes hilar adenopathy and CT scan of the neck notes cervical adenopathy. A biopsy of the left next mass leads to a diagnosis of a non-Hodgkins lymphoma. The Oncologist would like to proceed with CHOP and rituximab and is concerned about hepatitis C and B and would like recommendations

35. HCV RNA positive 6.8 log IU/ml
HCV genotype 1b
HBsAg negative
Anti-HBc positive, anti-HBs positive 35 mIU/mL
Ultrasound-unremarkable liver with suggestion of splenomegaly

36. treat HCV, along with lymphoma, with all oral non ribavirin containing directly acting anti-virals, and recommend anti- HBV prophylaxis.
 treat HCV, along with lymphoma, with all oral non ribavirin containing directly acting anti-virals and monitor HBV DNA for HBV flare
Treat HCV first as concurrent therapy of lymphoma may cause hematologic abnormalities and compromise HCV therapy
treat HCV, along with lymphoma, with all oral non ribavirin containing directly acting anti-virals; no need for HBV prophylaxis

37. Prevention of HBVr

38. Prevention of hepatitis B Flare

39. GRADE profile: antivirals vs. none
Summary of findings
Anticipated absolute effects
Partici-pants (studies)
GRADE rating categories
Placebo (monito-ring)
Antiviral prophylaxis
Relative effects
HBVrwith no Rx
Risk diff. w/ antiviral prophylaxis
Quality of evidence
Outcome: HBV reactivation (critical for decision making)
276
(5 RCTs)
12-18 mo f/u
Indirect-ness
50/137
36.5%
5/139
3.6%
RR 0.13
95% CI:0.6 – o.3
Low risk:
1 HBVr per 1,000
1 fewer HBVr per 1,000
[-1; -1]
 due to indirect-ness
Moderate:
50 HBVr per 1,000
44 fewer HBVr per 1,000
[-35; -47]
High risk:
500 HBVr per 1,000
435 fewer HBVr per 1,000
[-350; -470]
Left out: outcomes hepatitis (similar results); chemo disruption (no data); mortality (no data)

40. Antivirals with high barrier to resistance vs. lamivudine
Summary of findings
Anticipated absolute effects
Partici-pants (studies)
GRADE rating categories
With lamivudine
With
entecavir
Relative effects
Risk diff. with entecavir
Quality of evidence
Outcome: HBV reactivation (critical for decision making)
121
(1 RCTs)
40 mo f/u
Impreci-sion
18/60
30%
4/61
6.6%
RR 0.22
[0.08; o.6]
300 HBVr per 1,000
234 fewer HBVr per 1,000
[-117; -276]
 due to
imprecision
Outcome: HBV hepatitis flare (critical for decision making)
8/60
13.3%
0/61 0%
RR 0.06
[0.004; 1.0]
133 flares per 1,000
125 fewer flares per 1,000
[-133; -0]
Outcome: HBV related chemotherapy disruption (critical for decision making)
11/60
18.3%
1/61
1.6%
RR 0.09
[0.01; 0.67]
183 chemo disruptions per 1,000
167 fewer per 1,000 [-60; -181]
 due toimpreci-sion
Source: Huang et al. Presented at 2013 ASCO annual meeting. Data taken from slide set.

41. Recs: Screening for HBV *Screen using AASLD, EASL or APASL Guidelines
Patients at moderate or high risk: The AGA recommends screening for HBV (HBsAg and anti-HBc, followed by a sensitive HBV DNA test if positive) in patients who will undergo immunosuppressive drug therapy. GRADE: Strong recommendation, moderate quality evidence.
Patients at low risk: The AGA suggests against routinely screening for HBV in patients who will undergo immunosuppressive drug therapy and are at low risk for HBVr. GRADE: Weak recommendation, moderate quality evidence.
Comments: Patients belonging to populations with a baseline prevalence likely exceeding 2% for chronic HBV should be screened according to CDC and USPSTF recommendations.

42. Recs: antiviral prophylaxis (#1)
For patients at high risk (>10% risk): The AGA recommends antiviral prophylaxis over no prophylaxis in patients undergoing immunosuppressive drug therapy. GRADE: Strong recommendation, moderate quality evidence.
Comments: Treatment should be continued for at least 6 months after discontinuation of immunosuppressive therapy (at least 12 months for B-cell depleting agents).
Populations
Everyone receiving B-cell depleting agents who are core positive
HBsAg positive:
Anthracyclin derivatives
Moderate to high dose corticosteroids for ≥ 4 weeks

43. Recs: antiviral prophylaxis (#2)
For patients at moderate risk (1% - 10% risk): The AGA suggests antiviral prophylaxis over monitoring in patients undergoing immunosuppressive drug therapy. GRADE: Weak recommendation, moderate quality evidence.
Comments: Treatment should be continued for 6 months after discontinuation of immunosuppressive therapy. Patients who place a higher value on avoiding long-term use of antiviral therapy and the cost associated with its use and a lower value on avoiding the small risk of reactivation (particularly in those who are HBsAg negative) may reasonably select no prophylaxis over antiviral prophylaxis.
Populations
HBsAg positive or anti-HBc positive: TNF-alpha inhibitors, cytokine and integrin inbitors, tyrosine kinase inhibitors
HBsAg positive: low dose, long-term steroids
HBs Ag negative, anti-HBc positive: anthracyclin derivatives

44. Recs: antiviral prophylaxis (#3)
For patients at low risk (<1% risk): The AGA suggests against routinely using antiviral prophylaxis in patients undergoing immunosuppressive drug therapy at low risk for HBVr. GRADE: Weak recommendation, moderate quality evidence.
Populations
HBsAg positive or anti-HBc positive: Traditional immunosuppressive agents (e.g., azathioprine, 6-MP, methotrexate); intra-articular steroids; any doses of oral steroids daily for ≤ 1 week.
HBs Ag negative, anti-HBc positive patients treated with low-dose (< 10 mg prednisone or equivalent) corticosteroids for ≥ 4 weeks.

45. The (non) role of anti-HBs
Does the presence of anti-HBs in addition to anti-HBc in HBsAg-negative patients confer additional protection against HBVr?
Studies have shown pre-treatment anti-HBs titers to be an independent risk factor for reactivation only with rituximab (OR 0.003, 95%CI 0-0.3, p = )
No RCTs available that compare a monitoring strategy based on anit-HBs vs. antiviral prophylaxis to be non-inferior
Recommendation:
The AGA suggests against using anti-HBs status to guide antiviral prophylaxis for all risk groups. GRADE: Weak recommendation; Very low-quality evidence

46. Recs: choice of drugs (1)
The AGA suggests use of antiviral drugs with a high barrier to resistance over lamivudine for prophylaxis in patients undergoing immunosuppressive drug therapy. GRADE: Weak recommendation; Moderate quality evidence
Comments: Given the geographic variability in cost of antiviral therapy, those patients who put a higher value on cost and a lower value on avoiding the potentially small risk of resistance development (particularly in those who have an undetectable viral load and who are expected to use antiviral prophylaxis for ≤ 6 months) may reasonably select the least expensive antiviral hepatitis B medication over more expensive antiviral drugs with a higher barrier to resistance.

47. Anticipated absolute effects
Antivirals with high barrier to resistance vs. lamivudine in established HBV reactivation
Summary of findings
Anticipated absolute effects
Partici-pants (studies)
GRADE rating categories
With lamivudine
With anti-virals with high barrier
Relative effects
Quality of evidence
Outcome: Failure of virologic response (critical for decision making)
2,278
(7 RCTs)
52 weeks
Indirect-ness
533/1,106
48.2%
285/1,172
24.3%
RR 0.42
[0.3; o.59]
482 failures per 1,000
280 fewer failures per 1,000
[-198; -337]
 due to indirect-ness
Outcome: viral resistance development at 5 years (critical for decision making)
183
(1 RCT)
5 years NR
74%
1/183
0.55% (entecavir)
RR >100
740 cases of viral resistance per 1,000
735 fewer cases per 1,000
Source: Liu et al.2011; Woo G et al. 2010; Petersen et al. 2012; Chang TT et al. 2010; Kitrinos KM et al

48. Recs: choice of drugs (2)
The AGA recommends use of antiviral drugs with a high barrier to resistance over lamivudine for established HBVr in patients undergoing immunosuppressive drug therapy. GRADE: Strong recommendation; Moderate quality evidence

49. High Risk (reactivation risk > 10%)
AGA Institute Guidelines on Hepatitis B Reactivation ( HBVr) Clinical Decision Support Tool
High Risk (reactivation risk > 10%)
HBsAg-positive/anti-HBc-positive or HBsAg-negative/anti-HBc-positive
HBsAg-positive/anti-HBc-positive
Patients taking B cell depleting agents (e.g., rituximab, ofatumumab)
Patients taking anthracyclinederivatives (e.g., doxorubicin, epirubicin)
Patients taking moderate dose
(10-20 mg prednisone daily or equivalent ) or high dose ( > 20 mg prednisone daily or equivalent) corticosteroids daily for ≥4 weeks
Antiviral prophylaxis for at least 12 months after discontinuation of immunosuppressive therapy
Antiviral prophylaxis for at least 6 months after discontinuation of immunosuppressive therapy
GRADE – Strong Recommendation, moderate quality of evidence.
Reddy KR et al Gastroenterology ;148(1):215-9

50. Moderate Risk (reactivation risk 1-10%)
AGA Institute Guidelines on Hepatitis B Reactivation ( HBVr) Clinical Decision Support Tool
Moderate Risk (reactivation risk 1-10%)
HBsAg-positive/anti-HBc-positive or HBsAg-negative/anti-HBc-positive
HBsAg-positive/anti-HBc-positive
HBsAg-negative/anti-HBc-positive
Patients taking TNF alpha inhibitors (e.g., etanercept, adalimumab,
certolizumab, infliximab)
Patients taking other cytokine or integrin inhibitors (e.g., abatacept, ustekinumab, natalizumab, vedolizumab)
Patients taking tyrosine kinaseinhibitors (e.g., imatinib, nilotinib )
Patient taking low-dose
( < 10 mg prednisone daily or equivalent) corticosteroids daily for duration of ≥4 weeks
Patients taking moderate dose ( mg prednisone daily or equivalent ) or high dose ( > 20 mg prednisone daily or equivalent) corticosteroids daily for ≥4 weeks. Patients taking anthracycline derivatives (e.g., doxorubicin, epirubicin)
Suggest antiviral prophylaxis for at least 6 months after discontinuation of immunosuppressive therapy*
GRADE – Weak Recommendation, moderate quality of evidence
* Patients who place a higher value on avoiding the long-term use of antiviral therapy and cost associated with its use and a lower value on avoiding the small risk of reactivation (particularly in those who are HBsAg-negative), may reasonably select no prophylaxis over antiviral prophylaxis
Reddy KR et al Gastroenterology ;148(1):215-9

51. Low Risk (reactivation risk <1%)
AGA Institute Guidelines on Hepatitis B Reactivation ( HBVr) Clinical Decision Support Tool
Low Risk (reactivation risk <1%)
HBsAg-positive/anti-HBc-positive or HBsAg-negative/anti-HBc-positive
HBsAg-negative/anti-HBc-positive
Patients taking traditional immunosuppressive agents (e.g., azathioprine, 6-mercaptopurine, methotrexate)
Patients taking intra-articular corticosteroids.
Patients taking any dose oral corticosteroids daily for duration of ≤ 1 week
Patients taking low dose ( < 10 mg prednisone or equivalent) corticosteroids for ≥ 4 weeks
Suggest not to use routine antiviral prophylaxis in patients undergoing immunosuppressive drug therapy and are at low risk for HBVr
GRADE – Weak Recommendation
Moderate quality of evidence
Reddy KR et al Gastroenterology ;148(1):215-9

52. Summary
HBVr occurs with high frequency in patients undergoing immunosuppressive drug therapy
Now reported with HCV HBV: HCV DAA therapy
HBVr may lead to hepatitis, liver failure, liver transplant and death
Screening for HBV is recommended with 3 tests
Anti-HBc, anti-HBs and HBsAg
HBV DNA has a role in risk stratification
If HBsAg negative and HBV DNA +, treat as if HBsAg(+)
HBVr prophylaxis is highly effective and recommended in high risk populations
Patient’s values and preferences and trade-offs have to be taken into considerations when risk for HBVr is moderate

53. aknowlgements Raj Reddy for help with the slides and his knowledge
Robert Perrillo and all of his work on HBVr
AGA for supporting our project