Anabolic steroid-associated liver injury

Abstract

INTRODUCTION

Anabolic-androgenic steroids (AAS) are a class of chemical compounds that share the anabolic (myotrophic protein synthesis) and androgenic (male reproductive tract and secondary sexual characteristic development) activity of their parent steroid, testosterone. ¹ In addition to therapeutic indications such as hypogonadism, anemia, and muscle wasting or failure to thrive, AAS are illicitly consumed for strength training, muscle building, or euphoric effects. The AAS steroid structure has been the target of many chemical modifications aiming to optimize the anabolic-androgenic ratio, improve oral bioavailability, and mitigate metabolic inactivation. These numerous AAS chemical derivatives are often available illicitly or as dietary and herbal supplements, which frequently contain unapproved or prescription-only chemical compounds. ² AAS use has increased globally throughout the 21st century, with a majority of users reporting recreational or noncompetitive use. ³

MECHANISM OF ACTION OF AAS

AAS is characterized by its ability to exert androgenic and anabolic effects relative to testosterone, which are mediated through the intracellular androgen steroid receptor (AR). The differential ratio of anabolic versus androgenic activity is driven by tissue-specific receptor density and metabolism through 5α reductase, which converts testosterone into the more potent androgen dihydrotestosterone. ⁴

Anabolic effects occur in skeletal muscle tissue, which lacks 5α reductase, through the induction of muscle fiber hypertrophy, increase in progenitor cells, and promotion of myogenic differentiation, leading to increased muscle strength and mass.^4,5 The dose-response relationship between testosterone and skeletal muscle growth plateaus at supraphysiologic concentrations, suggesting that AAS may also exert anabolic action through activity at glucocorticoid receptors. ⁶

ROLE OF STRUCTURAL MODIFICATIONS TO AAS

Numerous chemical modifications to testosterone’s structure have been made to improve oral bioavailability by inhibiting hepatic first-pass metabolism, prolonging the duration of action, and optimizing the anabolic-androgenic activity ratio. Complete dissociation of the desirable anabolic effects from the undesirable androgenic effects has not been achieved. ⁶ Testosterone is further metabolized by 5α reductase to produce dihydrotestosterone, which is resistant to inactivating conversion to estradiol by aromatase and is thus a more potent androgen than the parent testosterone. Subsequent reduction of the 3-keto group and oxidation of a 17β hydroxy group to a 17-keto metabolite occur rapidly and are important for urinary elimination of inactive metabolites. ¹ Clinically relevant structural modifications target these metabolic steps, such as alkylation at C7 to prevent reduction by 5α reductase, 17α alkylation to reduce intrahepatic metabolism and confer oral bioavailability, or 17β esterification to prolong the duration of action.^1,4

AAS ASSOCIATED DRUG-INDUCED LIVER INJURY (DILI)

Hepatic adverse effects are classically associated with 17α-alkylated AAS due to their ability to bypass first-pass hepatic metabolism. Historical reporting of hepatotoxicity from AAS may be overestimated due to the association of bodybuilding and strength training with elevations in alanine transaminase (ALT) and aspartate transaminase independent of AAS use.^1,7 Adverse effects of AAS, other than DILI, are summarized in Table 1.

TABLE 1.

Notable adverse effects of AAS^4,5

Organ system	Notable adverse effects
Behavioral	Aggression Anxiety Depression Euphoria Mood lability Mania
Cardiovascular	Cardiac arrhythmias Dyslipidemia Myocardial hypertrophy
Dermatologic	Acne Alopecia Hirsutism
Hematologic	Erythrocytosis Thrombosis
Hepatic	Benign liver lesions: diffuse hyperplasia, focal nodular hyperplasia, nodular regenerative hyperplasia Cholestatic jaundice Elevated liver enzymes Neoplastic lesions: hepatic adenoma, HCC Peliosis hepatitis: proliferation of sinusoidal hepatic capillaries leading to numerous intrahepatic hemorrhagic cysts
Renal	Acute kidney injury Renal failure
Reproductive	Gynecomastia Infertility Libido changes Prostatic hypertrophy

The clinical course of AAS DILI is described in several small case series and individual case reports. Prevalence data from the Spanish and Latin American DILI network reflect the growing importance of AAS DILI, from 1% of DILI reports in 1994 to 8% of total reports by 2013. ⁸

Summaries of AAS DILI clinical phenotypes are shown in Table 2. Patients are likely to be male, in their 30s, and without contributing medical history. Common presenting symptoms include jaundice, pruritus, lethargy, abdominal discomfort, and/or anorexia with significant weight loss. Typical laboratory values at presentation include conjugated hyperbilirubinemia, mildly elevated alkaline phosphatase (ALP), normal or minimally elevated gamma-glutamyl transpeptidase, and mild elevations in ALT or aspartate transaminase.^8–11 ALT and aspartate transaminase rapidly normalize after presentation. ¹⁰ The R-value, the ratio of ALT to ALP (relative to normal values) commonly used to characterize enzyme patterns of liver injury, usually demonstrates a hepatocellular pattern at presentation due to normal ALP. Normalization of ALT and lagging peak increase in ALP drive a shift in R-value to a mixed or cholestatic pattern of injury after presentation. This cholestatic pattern of injury can have a prolonged course with severe pruritus, and this consistent liver injury phenotype is unique among drugs or HDS, allowing for greater certainty with causality assessment.^8–11

TABLE 2.

Clinical phenotype of AAS DILI based on published case series^8–11

	Timeframe	AST/ALT	Bilirubin	ALP	GGT	R-value	Notes
Initial presentation	Variable 2–10 weeks from time of AAS consumption	↑ ALT > AST ALT commonly > 3 × ULN	↑↑ Median 9.8 mg/dL	↔ / ↑ Normal in 70% of patients	↔	Variable, but commonly hepatocellular pattern	Cholestatic pattern injury at presentation is more likely to be associated with severe hospital course and renal insufficiency
Peak	1–2 weeks after presentation	↑ AST > ALT Peak values similar to the magnitude to presenting values	↑↑↑ In severe cases, > 40 mg/dL	↑ Generally 2 × ULN	↔	Cholestatic pattern (downtrend of R-value from presentation)	Peak total bilirubin> 40 mg/dL associated with renal insufficiency AAS DILI cases have higher peak bilirubin and lower peak ALP/ALT compared to other DILI cases

Note: R-value = (ALT/ULN ALT)/(ALP/ULN ALP)

↔=value within normal limits

↑=value 2–5 × ULN

↑↑=value 5-10 × ULN

↑↑↑=value>10 × ULN

Abbreviations: ALP, alkaline phosphatase; ALT, alanine transaminase; AST, aspartate transaminase; GGT, gamma-glutamyl transpeptidase; ULN, upper limit of normal.

Predictors of severe clinical course include the cholestatic pattern of injury at presentation and significantly elevated peak serum total bilirubin (> 40 mg/dL).^8–10 Renal dysfunction occurred in ~20% of reported cases, with all experiencing renal recovery.^9,10 On liver biopsy, the most common histology seen is bland intralobular cholestasis or mixed hepatocellular and cholestatic injury. Bile duct injury is rare, and no bridging or confluent necrosis has been reported.^9,10,12,13 Most series have not reported either liver transplants or deaths from AAS DILI, although an Australian series did report a liver transplant for 1 patient with ductopenia, a histologic feature not typically seen with AAS DILI, and a single death was reported in a Latin American series without specific details.^8,9,12,13 Coagulopathy (international normalized ratio > 1.5) occurs in less than 15% of cases, which may be due to decreased absorption of vitamin K rather than liver synthetic dysfunction.^8,10 Encephalopathy has only been reported in a single case but occurred concomitantly with renal failure, suggesting multifactorial etiology. ¹¹ There is limited evidence for drug treatment of AAS DILI, with a US case series reporting rare use of ursodiol (50%) and corticosteroids (16%) without any improvement in liver enzymes. ¹⁰ Normalization of enzymes and jaundice has been demonstrated at a median of 3 months, with nearly all patients having normal ALT at 1 year.^9,10 In recent AAS DILI cases, oral is the most common route of administration, supplanting intramuscular stanozolol as the most commonly implicated agent.^8,10 In a US case series utilizing chemical analysis of patient-submitted AAS products, 52% contained AAS that were different from those agents listed on the product label and only 18% had confirmation of the AAS listed on the product label. ¹⁰

As AAS DILI demonstrates a similar clinical phenotype of impaired bile salt excretion as seen in patients with mutations in genes associated with progressive familial intrahepatic cholestasis syndromes, mutations in those genes have been reported in some individuals with AAS DILI. Genotyping in a US case series demonstrated 7 of 44 patients having damaging missense mutations in those genes associated with progressive familial intrahepatic cholestasis syndromes (ABCB11, ATP8B1, and ABCB4). ¹⁰ Similarly, immunohistochemical staining of liver biopsy specimens demonstrates deficient canalicular expression of transporters associated with bile salt excretion despite the absence of pathogenic mutations identified in ATP8B1 and ABCB11 genes, suggesting impaired processing of these proteins. ⁹

Detection of AAS use is challenging, as AAS may be unknowingly present in supplements (either unlabeled or mislabeled) or may be acquired illicitly and thus unreported by patients. Adverse effects such as infertility, gynecomastia, sexual dysfunction, alopecia, acne, increased muscularity, and testicular atrophy may serve as clues to their use. Laboratory testing, including elevated hemoglobin and hematocrit and suppressed serum luteinizing hormone, follicle-stimulating hormone, or testosterone levels, may also be clues to AAS use. ⁷ The epi-testosterone to testosterone ratio and urinary C13/C12 ratio of androgenic steroids have also been shown to be useful for the detection of AAS consumption. ¹⁰ Urine tests utilizing mass spectrometry can detect many AAS. The use of pure testosterone can be more difficult to detect, although a testosterone-to-estrogen ratio > 4 suggests exogenous administration. ⁷

ROLE OF REGULATORY OVERSIGHT

In the United States, AAS are classified as Schedule III controlled substances by the Drug Enforcement Agency (DEA) following the “Anabolic Steroid Control Act” of 2004, which includes language designed for the broad inclusion of novel AAS derivatives. Periodic updates to the legislature explicitly list new chemical derivatives under this regulatory umbrella. ¹⁴

Many AAS elude these regulations as ingredients in supplements that may be adulterated with controlled substances, approved pharmaceutical products, or are otherwise improperly labeled. As a consequence, AAS are included in dietary and herbal supplements either intentionally or unintentionally (through cross-contamination or poor-quality manufacturing) and are not always listed on the product label, even when intentionally included. ¹⁴ AAS thus poses dangers to patients, clinicians, and regulators.

CONCLUSION

AAS are illicitly consumed to increase muscle mass, enhance athletic performance, and have euphoric effects. AAS activity is balanced between undesirable androgenic and desirable anabolic effects. There have been many attempts at chemical substitutions to the AAS structure with the goal of optimizing this ratio and improving pharmacokinetic properties such as oral bioavailability. AAS may be consumed intentionally or unintentionally as components of dietary supplements or otherwise acquired illicitly. The phenotype of DILI related to AAS is predominantly cholestatic with jaundice, with occasional mixed hepatocellular injury seen on liver biopsy. AAS DILI is associated with prolonged and significantly elevated serum total bilirubin, in contrast to DILI associated with other drugs or herbal supplements. In rare cases, patients can experience renal injury, which is presumed to be related to their high serum bile acid levels. Supportive care is critical to the clinical care of these patients, as corticosteroids or ursodiol have not been reported to improve clinical course. The index of suspicion for AAS consumption should remain high in otherwise healthy patients who present with prolonged cholestatic jaundice with a standard liver injury evaluation that is otherwise negative.