Liver Safety Assessment in Clinical Trials of New Agents for Chronic Hepatitis B

Robert J. Fontana; Mark I. Avigan; Harry L. A. Janssen; Arie Regev; Poonam Mishra; Anuj Gaggar; Nathaniel Brown; Cynthia Wat; Patricia Mendez; Ryan T. Anderson; Bruce Given; Veronica Miller; Maria Beumont

Disclosures

J Viral Hepat. 2020;27(2):96-109.

Conclusions

Various laboratory, clinical and histological criteria should be considered in the design of future phase 1 and 2/3 studies of investigational regimens for chronic HBV (Table 3). To adjudicate serum ALT flares encountered in these drug development programmes, a predefined protocol that specifies the testing of liver safety biomarkers prior to and during treatment should be established and followed (Table 4 and Table 5). Furthermore, investigation of individual cases of liver biomarker safety signals using a comprehensive and methodical approach is advisable. When a potential pattern of DILI events is identified in trials of new anti-HBV agents, an expert adjudication panel is recommended to help assess the overall preclinical, clinical and pharmacological liver safety data. Finally, molecular diagnostic assays that can reliably differentiate an enhanced host immune response to HBV antigens vs drug hepatotoxicity are in development and will hopefully improve our ability to develop safe and effective medications that increase the rate of HBsAg loss in the millions of patients with chronic HBV worldwide.

1 2 3 4 5 6 7 8 9 10

Next Section

Funding information
MEDIAN Technologies; Gilead Sciences; Hepatitis B Foundation; Arbutus Biopharma; Aicuris; GlaxoSmithKline; Abbott Molecular; Altimmune; DDL Diagnostics; ContraVir; Astra Zeneca; Springbank Pharma; Novartis; Assembly Biosciences; Arrowhead

Abbreviations
AFP, Alpha fetoprotein; ALF, Acute liver failure; ALK, Alkaline phosphatase; ALT, Alanine aminotransferase; Anti-HBe, Antibody to hepatitis B e antigen; Anti-HBs, Antibody to hepatitis B surface antigen; AST, Aspartate aminotransferase; BMI, Body mass index; cccDNA, Covalently closed circular DNA; CMV, Cytomegalovirus; CPK, Creatinine phosphokinase; DILI, Drug-induced liver injury; EBV, Epstein-Barr virus; FDA, Food and Drug Administration; GWAS, Genome-wide association study; HBeAg, Hepatitis B e antigen; HBsAg, Hepatitis B surface antigen; HBV, Hepatitis B virus; HCC, Hepatocellular carcinoma; HCV, Hepatitis C virus; HDS, Herbal and dietary supplements; HDV, Hepatitis D virus; HEV, Hepatitis E virus; HIV, Human immunodeficiency virus; NAFLD, Nonalcoholic fatty liver disease; NrtI, Nucleos(t)ide reverse transcriptase inhibitors; PD, Pharmacodynamic; PK, Pharmacokinetic; PROD, Paritaprevir, ritonavir, ombitasvir and dasabuvir; RUCAM, Rousell Uclaf Causality assessment method; ULN, Upper limit of normal.

Disclaimer
The views expressed are those of the authors and do not necessarily represent the position of, nor imply endorsement from, the US Food and Drug Administration or the US Government.

Table 1. Types of serum ALT flares observed in Hepatitis B patients
Flare type	Timing & characteristics	Clinical sequelae and actions
Host mediated	Spontaneous- enhanced host immunity to infected hepatocytes; frequently preceded by surge in HBV replication; variable ALT Treatment related-enhanced host immunity to infected hepatocytes	HBeAg (or HBsAg) loss in some; severe flare may require rescue NRTI Early flare associated with ↓ HBV replication; continue therapy if no ↑ Bili or INR HBeAg (or HBsAg) loss in some; continue therapy if no ↑ Bili or INR
	Early <12 wk; variable ALT
	Late >12 wk; variable ALT
Virally mediated	On-Treatment - redetection of previously suppressed HBV-DNA	Non-compliance associated with resurgence of wild-type HBV; may respond to resumption of Rx Drug resistant breakthrough associated with viral variants Severe may require rescue NRTI
	Late >12 wk; variable ALT
	Post-treatment: redetection of HBV-DNA within 48 wk of therapy completion
Idiosyncratic drug toxicity	Timing: may occur at any time; independent of drug dose or other host factors Phenotype: Variable ALT; some with ↑Alk phos or bili	Variable phenotype makes diagnosis difficult Serum ALT >10× ULN or ↑T. bili or INR require immediate drug d/c Potentially severe in those with advanced fibrosis/cirrhosis

Table 2. Incidence of serum ALT elevations in Hepatitis B patients receiving oral NrtIs and PegIFN in registration trials ^a
Tenofovir disproxil fumarate
Maximum ALT elevation		Number with abnormality/number tested (%)
Maximum ALT elevation		HBeAg (+) N = 524	HBeAg (−) N = 441
>1 to ≤3× ULN	Baseline	295 (56.3)	276 (62.6)
	On-treatment: ≤12 wk	262 (50.0)	254 (57.6)
	Year 1	290 (55.3)	205 (46.5)
	Year 2	198 (37.8)	139 (31.5)
>3 to ≤5× ULN	Baseline	107 (20.4)	61 (13.8)
	On-treatment: ≤12 wk	105 (20.0)	49 (11.1)
	Year 1	39 (7.4)	10 (2.3)
	Year 2	15 (2.9)	7 (1.6)
>5 to ≤10× ULN	Baseline	68 (13.0)	47 (10.7)
	On-treatment: ≤12 wk	77 (14.7)	24 (5.4)
	Year 1	13 (2.5)	0 (0.0)
	Year 2	2 (0.4)	3 (0.7)
>10× ULN	Baseline	22 (4.2)	8 (1.8)
	On-treatment: ≤12 wk	35 (6.7)	8 (1.8)
	Year 1	2 (0.4)	0 (0.0)
	Year 2	1 (0.2)	1 (0.2)
Peginterferon alpha-2a
Maximum ALT elevation		Number with abnormality/number tested (%)
Maximum ALT elevation		HBeAg (+) N = 481	HBeAg (-) N = 330
>1 to ≤3× ULN	Baseline	229 (47.6)	217 (65.8)
	On-treatment: ≤12 wk	175 (36.4)	136 (41.2)
	On-treatment (48 wk)	149 (31.0)	134 (40.6)
	Post-treatment^b	188 (39.1)	132 (40.0)
>3 to ≤5× ULN	Baseline	117 (24.3)	80 (20.7)
	On-treatment: ≤12 wk	131 (27.2)	73 (22.1)
	On-treatment (48 wk)	129 (26.8)	77 (23.3)
	Post-treatment^b	55 (11.4)	42 (12.7)
>5 to ≤10× ULN	Baseline	89 (18.5)	49 (12.7)
	On-treatment: ≤12 wk	120 (25.0)	58 (17.6)
	On-treatment (48 wk)	139 (28.9)	68 (20.6)
	Post-treatment^b	67 (13.9)	32 (9.7)
>10× ULN	Baseline	22 (4.57)	10 (2.6)
	On-treatment: ≤12 wk	43 (8.9)	26 (7.9)
	On-treatment (48 wk)	57 (11.9)	32 (9.7)
	Post-treatment^b	43 (8.9)	31 (9.4)

^aData obtained from registration trials of Peg-IFNa2a and TDF that enrolled immune active hepatitis B patients with elevated ALT levels at baseline. The absolute frequency of ALT elevations has been reported for each period with no analysis in comparison with the individual patient's baseline level (i.e no assessment of treatment-emergent ALT elevations).
^bPost-treatment period of 24 wk.

Table 3. Recommended exclusion criteria and laboratory monitoring in chronic hepatitis B Clinical Trial patients
Topic	Exclusion criteria^a
Other causes of liver disease	- Anti-HCV (+) with detectable HCV RNA, anti-HDV (+), or anti-HIV (+) - Evidence of other causes of liver injury via liver biopsy, serological tests, liver imaging (ultrasound, CT scan or MRI), or medical history^b
Known or suspected hepatocellular carcinoma (HCC)	- Prior history of HCC - Imaging study demonstrating solid mass c/w HCC - Serum AFP > 100 ng/mL independent of liver imaging results - If serum AFP > ULN but <100, must have CT/MRI with contrast showing no evidence of HCC
Liver disease severity	Phase 1–2: Advanced fibrosis/cirrhosis (any CTP score) established by any of the following: - Liver biopsy showing cirrhosis - Imaging evidence of suspected cirrhosis - Fibroscan stiffness score >9–12 kPa^c Laboratory evidence suggestive of advanced disease including platelets < 100 000/mL, T Bili > 1.5× ULN, albumin < 3.4 g/dL or INR > 1.2 Phase 3: - CTP B or C cirrhosis - T Bili > 2× ULN
Serum ALT, Alk P, HBV DNA	NRTI suppressed: - Serum ALT > 3× ULN or >120 U/L, or - ALK > 2× ULN, or - HBV DNA > LLD within 6 mo of screening Treatment naïve or nonresponders: - Serum ALT > 7–10× ULN or >300 U/L, or - ALK > 2× ULN
Study phase	Recommended laboratory monitoring
Phase 1 and 2a	• Monitor liver safety biomarkers (serum ALT, AST, ALK, T Bili, INR and albumin) every 1–2 wk during first month of treatment, monthly thereafter on-treatment and at week 4, 8, 12, 24 and 48 after dosing
Phase 2b and 3	• Monitor liver safety biomarkers every 2–4 wk in the first 3 mo and every 4–8 wk thereafter during dosing and at week 4, 8, 12, 24 and 48 after dosing • In drugs with suspected hepatotoxicity^d, monitor liver safety biomarkers every 2–4 wk for the first 6 mo and every 4–8 wk thereafter

^aSpecific exclusion criteria may vary based upon the mechanism of action and safety information from preclinical and early clinical studies.
^bFor example, tests to exclude genetic hemochromatosis, autoimmune hepatitis, and alcoholic liver disease.
^cThe fibroscan cut-off value for cirrhosis is lower in obese patients.
^dBased upon preclinical data, available clinical trial data, and drug pharmacokinetics/mechanism of action AFP, alpha fetoprotein; ALT, alanine aminotransferase; CTP, Child-Turcotte-Pugh; DILI, drug induced liver injury HCC, hepatocellular carcinoma; INR, international normalized ratio; LLD, lower limit of detection; PLT, platelets; T Bili, total bilirubin.

Table 4. Recommended management of liver safety signals in clinical trials of NRTI suppressed Hepatitis B patients
Treatment emergent ALT elevation	Treatment emergent total/direct bilirubin elevationa	Liver-related symptoms	Action
Normal baseline: ALT ≥5× ULN Elevated baseline ^b: ALT ≥3× baseline (or ≥3× new nadir^c) or ≥300 U/L (whichever occurs first)	Normal Patients with Gilbert's syndrome: No change in baseline TBL	None	Repeat ALT, AST, ALP, TBL, INR, albumin, obtain HBV DNA^d within 3–5 d and initiate close monitoring^e until levels return toward nadir values Evaluate for etiology of ALT elevation (Table 6)
Normal baseline: ALT ≥8× ULN Elevated baseline ^b: ALT ≥5× baseline (or ≥5× new nadir^c) or ≥500 U/L (whichever occurs first)	Normal Patients with Gilbert's syndrome: No change in baseline TBL	None	Repeat ALT, AST, ALP, TBL, INR, albumin and obtain HBV DNA^d in 2–3 d. If abnormality persists, consider interrupting study drug and initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 6). Study drug can be restarted only if a self-limited non-drug etiology is identified
Normal baseline: ALT ≥3× ULN Elevated baseline ^b: ALT ≥2× baseline (or ≥2× new nadir^c) or ≥300 U/L (whichever occurs first)	Normal Patients with Gilbert's syndrome: No change in baseline TBL	Severe fatigue, nausea, vomiting, right upper quadrant pain	Immediately interrupt study drug^f. Repeat ALT, AST, ALP, TBL, INR, albumin and obtain HBV DNA^d in 2–3 d. Initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 6). Study drug can be restarted only if a self-limited non-drug etiology is identified
Normal baseline: ALT ≥3 ULN Elevated baseline ^b: ALT ≥3× baseline (or ≥3× new nadir^c) or ≥300 U/L (whichever occurs first)	TBL ≥2× ULN Patients with Gilbert's syndrome: Doubling of direct bilirubin	None	Immediately interrupt study drug^f. Repeat ALT, AST, ALP, TBL, INR, albumin and obtain HBV DNA^d in 2–3 d Initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 8). Study drug can be restarted only if a self-limited non-drug etiology is identified

Notes: Modified from reference 45.
Abbreviations: Alb, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBL, total bilirubin; ULN, upper limit of normal.
^aTotal bilirubin and direct bilirubin fractionation is recommended to help identify patients with indirect hyperbilirubinemia due to gilbert's syndrome and hemolysis versus liver injury.
^bElevated baseline is defined as ALT ≥1.5× ULN.
^cIn patients with a stable decrease in ALT during treatment (>50% of baseline value), a new baseline, corresponding to the ALT nadir, should be established as reference for subsequent determination of a DILI signal.
^dAdditional recommended HBV-related tests: quantitative hepatitis B surface antigen (HBsAg quant), quantitative hepatitis B e antigen (HBeAg quant), quantitative hepatitis B surface antibody (anti-HBs quant).
^eFrequency of monitoring may need to be adjusted based on clinical scenario and severity of injury.
^fDrug interruption and discontinuation decisions should involve assessment for other causes of abnormal hepatic biochemical tests including HBV reactivation as well as the half-life and dosing interval of the investigational agent (see Table 6). In some instances, NRTIs may be continued in patients receiving these agents in combination with an investigational drug.

Table 5. Recommended management of liver safety signals in clinical trials of treatment naive or prior nonresponder hepatitis B patients
Treatment emergent ALT elevation	Treatment emergent total/direct bilirubin elevation^f	Liver-related symptoms	Action
Normal baseline: ALT ≥5× ULN Elevated baseline ^b: ALT ≥3× baseline (or ≥3× new nadir^c) or ≥500 U/L (whichever occurs first)	Normal Patients with Gilbert's syndrome: No change in baseline TBL	None	Repeat ALT, AST, ALP, TBL, INR, albumin, obtain HBV DNA^d within 3–5 d and initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 6)
Normal baseline: ALT ≥8× ULN Elevated baseline ^b: ALT ≥5× baseline (or ≥5× new nadir^c) or ≥800 U/L (whichever occurs first)	Normal Patients with Gilbert's syndrome: No change in baseline TBL	None	Repeat ALT, AST, ALP, TBL, INR, albumin and obtain HBV DNA^d in 2–3 d. If abnormality persists, consider interrupting study drug and initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 6). Study drug can be restarted only if a self-limited non-drug etiology is identified
Normal baseline: ALT ≥3× ULN Elevated baseline ^b: ALT ≥2× baseline (or ≥2× new nadir^c) or ≥500 U/L (whichever occurs first)	Normal Patients with Gilbert's syndrome: No change in baseline TBL	Severe fatigue, nausea, vomiting, right upper quadrant pain	Immediately interrupt study drug^f. Repeat ALT, AST, ALP, TBL, INR, albumin and obtain HBV DNA^d in 2–3 d. Initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 6). Study drug can be restarted only if a self-limited non-drug etiology is identified
Normal baseline: ALT ≥3 ULN Elevated baseline ^b: ALT ≥3× baseline (or ≥3× new nadir^c) or ≥300 U/L (whichever occurs first)	TBL ≥2× ULN Patients with Gilbert's syndrome: Doubling of direct bilirubin	None	Immediately interrupt study drug^f. Repeat ALT, AST, ALP, TBL, INR, albumin and obtain HBV DNA^d in 2–3 d Initiate close monitoring^e until levels return toward nadir values. Evaluate for etiology of ALT elevation (Table 6). Study drug can be restarted only if a self-limited non-drug etiology is identified

Notes: Modified from reference.⁴⁵
Abbreviations: Alb, albumin; ALP, alkaline phosphatase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBL, total bilirubin; ULN, upper limit of normal.
^aTotal bilirubin and direct bilirubin fractionation is recommended to help identify patients with indirect hyperbilirubinemia due to gilbert's syndrome and hemolysis versus liver injury.
^bElevated baseline is defined as ALT ≥1.5× ULN.
^cIn patients with a stable decrease in ALT during treatment (>50% of baseline value), a new baseline, corresponding to the ALT nadir, should be established as reference for subsequent determination of a DILI signal.
^dAdditional recommended HBV-related tests: quantitative hepatitis B surface antigen (HBsAg quant), quantitative hepatitis B e antigen (HBeAg quant), quantitative hepatitis B surface antibody (anti-HBs quant)
^eFrequency of monitoring may need to be adjusted based on clinical scenario and severity of injury.
^fDrug interruption and discontinuation decisions should involve assessment for other causes of abnormal hepatic biochemical tests including HBV reactivation as well as the half-life and dosing interval of the investigational agent (see Table 6). In some instances, NRTIs may be continued in patients receiving these agents in combination with an investigational drug.

Table 6. Recommended evaluation of Hepatitis B patients with unexplained serum liver biochemistry abnormalities in clinical trials.
Competing cause	Recommended evaluation^a	Interpretation
1st line testing
Liver directed medical history and physical exam	Recent travel/exposures Alcohol consumption Exercise & activity Concomitant medications & HDS product consumption	Consider HAV, HCV, HDV, HEV If excessive or AST/ALT >2 consider lab testing^a Possible rhabdomyolysis Drug hepatotoxicity and acetaminophen hepatotoxicity
Acute HAV	Anti-HAV (IgM)	Acute HAV infection
Acute HCV	Anti- HCV HCV RNA (PCR)	Parenteral exposure/risk factor Acute HCV may be anti-HCV (−) but HCV RNA (+)
Muscle injury	Excessive muscle use history Serum CPK, aldolase	Compare to baseline values, AST frequently elevated as well
Alcoholic liver damage	Urinary ethylglucuronide Serum Phosphatidylethanol	Alcohol use in past 3–5 d Alcohol use in past 3 wk
Pancreaticobiliary disease, HCC	Liver imaging such as ultrasound/CT or MRI^a	Evaluate for gallstones, pancreatitis, PV thromboses, malignancy If cholestatic, MRCP recommended
2nd line testing^b
Autoimmune hepatitis	ANA, SmAb Quantitative IgG, IgM, IgA	Compare to available baseline Liver biopsy needed to confirm a diagnosis of AIH
Hepatic ischemia	Review of blood pressure/pulse Electrocardiogram	Echocardiogram or cardiology consult may be indicated
Other hepatotoxins	Urine toxicology screen	Cocaine, opiates and other illicit substances may cause liver damage
Acute HDV	Anti-HDV	Selected patients at increased risk
Acute HEV	Anti-HEV IgM, IgG, and HEV-RNA^c	First line test in endemic areas Reference lab may be required for testing and HEV RNA confirmation
CMV, EBV, HSV infection	EBV-DNA, CMV- DNA, HSV-DNA by PCR	May need to obtain acute and convalescent serologies; liver biopsy needed to confirm HSV
Cholestasis of sepsis	Review of all medical records	Clinical diagnosis of exclusion

^aExtent and type of work-up may vary by patient location and flare severity; all patients with jaundice should undergo liver imaging.
^bThis testing should be undertaken in patients without an identified cause of acute liver injury after initial evaluation.
^cA reference lab with a rapid turnaround time that can confirm anti-HEV IgM positive samples with nested PCR should be considered.

Supplementary Table 1. Grading of serum ALT flares in untreated Chronic HBV patients.
Grade	Term	Serum ALT Level (ULN)
1	Minimal	>1x to ≤3x
2	Mild	>3x to ≤5x
3	Moderate	>5x to ≤10x
4	Marked or severe^a	>10x

^aAny serum ALT elevation with a total bilirubin >2.5 mg/dl or INR >1.5 or in patients with clinical symptoms suggestive of hepatitis are considered severe

Authors and Disclosures

Robert J. Fontana¹, Mark I. Avigan², Harry L. A. Janssen³, Arie Regev⁴, Poonam Mishra⁵, Anuj Gaggar⁶, Nathaniel Brown⁷, Cynthia Wat⁸, Patricia Mendez⁹, Ryan T. Anderson¹⁰, Bruce Given¹¹, Veronica Miller¹⁰ and Maria Beumont¹²

¹Division of Gastroenterology and Hepatology, University of Michigan, Ann Arbor, MI, USA
²Office of Surveillance and Epidemiology, Center for Drug Evaluation and Research, US Food and Drug Administration, Silver Spring, MD, USA
³Toronto Centre for Liver Disease, University Health Network, Toronto, Canada
⁴Eli Lilly and Company, Indianapolis, IN, USA
⁵Division of Antiviral Products, Center for Drug Evaluation and Research, US Food and Drug Administration, Silver Spring, MD, USA
⁶Gilead Sciences, Foster City, CA, USA
⁷Hepatitis B Foundation, Doylestown, PA, USA
⁸Roche Products, Welwyn Garden City, UK
⁹Arbutus Biopharma Inc., Warminster, PA, USA
¹⁰Forum for Collaborative Research, University of California, Berkeley
¹¹Arrowhead Pharmaceuticals, Inc, Pasadena, CA, USA
¹²Janssen Research & Development, Janssen Pharmaceutica NV, Beerse, Belgium

Correspondence
Robert J. Fontana, MD, Professor of Medicine, University of Michigan, 3912 Taubman, Ann Arbor, MI 48109. Email: rfontana@med.umich.edu

Present address
Patricia Mendez, Immunocore, Conshohocken, PA, USA
Ryan T. Anderson, Mathematica Policy Research, Washington, DC, USA

Disclosures
Robert Fontana: receives research funding from BMS, Gilead and Abbvie and has provided consulting for Alynam/Sanofi. Mark Avigan: reports nothing to disclose. Harry Janssen: has received grants from AbbVie, Bristol Myers Squibb, Gilead Sciences, Janssen, Medimmune, Merck and Roche, and is a consultant for AbbVie, Benitec, Bristol Myers Squibb, Gilead Sciences, Janssen, Medimmune, Merck, Roche, Arbutus, Vir Biotechnology Inc Arie Regev: is a full-time employee of Eli Lilly and owns stocks of Eli Lilly. Poonam Mishra: reports nothing to disclose. Anuj Gaggar: is an employee of Gilead Biosciences. Nathaniel Brown: consultancies with two small companies, one HBV-focused and one HCV-focused. Cynthia Wat: is an employee of Roche and has stockholdings in Roche. Patricia Mendez: is an employee of Mallinckrodt Pharmaceuticals. Ryan Anderson: reports nothing to disclose. Bruce Given: reports that he is an employee and shareholder of Arrowhead Pharmaceuticals, Inc Veronica Miller: reports support from Abbott Molecular, Aicuris, Altimmune, Arbutus Biopharma, Arrowhead Pharmaceuticals, Assembly Biosciences, Astra Zeneca, ContraVir, DDL Diagnostics, Gilead Sciences, GlaxoSmithKline, Hepatitis B Foundation, Janssen, MEDIAN Technologies, Novartis, PPD, Quest Diagnostics, Roche Molecular Systems and Springbank Pharma. Maria Beumont: is an employee of Johnson and Johnson.

Author Contributions
All the authors contributed to the manuscript design, drafting, figures and final review.