Assessment of Amikacin- and Capreomycin-Related Adverse Drug Reactions in Patients with Multidrug-Resistant Tuberculosis and Exploring the Role of Genetic Factors

{"Background": "Multidrug-resistant tuberculosis (MDR-TB) remains a significant public health problem, with 450,000 cases of rifampicin-resistant or MDR-TB reported globally in 2021. Before the introduction of all-oral regimens, second-line injectable agents such as amikacin (an aminoglycoside) and capreomycin (a polypeptide) were widely used in MDR-TB treatment. These drugs are associated with adverse drug reactions (ADRs), particularly ototoxicity (irreversible, bilateral high-frequency hearing loss or vestibular hypofunction) and nephrotoxicity (often reversible renal tubular damage). The incidence of these ADRs varies considerably across studies, and individual susceptibility is influenced by factors including dose, treatment duration, age, renal function, comorbidities, and potentially genetic factors\u2014particularly mitochondrial DNA (mtDNA) variants in the 12S rRNA gene (e.g., m.1555A>G, m.1494C>T). This study aimed to analyze the occurrence of amikacin- and capreomycin-related ADRs in Latvian MDR-TB patients and evaluate the role of patient-, disease-, and therapy-related factors, including full-length mtDNA sequencing.", "Methods": "This retrospective study included 47 adult MDR-TB patients who received amikacin and/or capreomycin between 2014 and 2017. Data on treatment regimens, comorbidities, adverse events, treatment outcomes, and doses received were collected. Ototoxicity was confirmed by clinical sensorineural hearing loss on audiogram, dizziness, vertigo, or tinnitus. Nephrotoxicity was defined as renal failure due to drug usage. Patient questionnaires captured age, sex, smoking, alcohol consumption, body weight, and height. Full mitochondrial genome sequencing was performed using IonTorrent technology (PGM). The whole mtDNA genome was amplified via two PCR reactions producing >8 kb fragments, which were sonicated into 200\u2013250 bp fragments. Sequencing data were analyzed using the Galaxy web platform, with reads aligned to the revised Cambridge Reference Sequence (rCRS, NC_012920.1). Single-nucleotide variants (SNVs) were identified using FreeBayes; indels were not analyzed. Haplogroups were determined using HaploGrep 2.0. Statistical analyses included Chi-square test, Fisher's exact test, logistic regression, ANOVA, and ANCOVA, with p \u2264 0.05 considered significant.", "Key Results": "Of the 47 patients (average age 48 years, range 20\u201387), 2 received amikacin only, 29 received capreomycin only, and 16 received both. Patients received an average of 109 doses (range 6\u2013346, SD=83). Overall, 55.3% (26/47) experienced ADRs: ototoxicity occurred in 34.0% (16/47), nephrotoxicity in 27.7% (13/47), and 6.4% (3/47) experienced both. All but one ototoxicity patient showed clinical hearing loss in high frequencies. Ototoxicity was significantly associated with amikacin use (logistic regression: p=0.049, Chi-Square 3.88, 95% CI: 0.002\u20130.915; specificity 90.3%, sensitivity 50.0%, 76.6% correct predictions). Ototoxicity was significantly less frequent in patients receiving capreomycin (p=0.004). No association was found between ototoxicity and number of doses, age, BMI, smoking, alcohol, comorbidities, or pre-existing hearing impairment. Nephrotoxicity was associated with older age (p=0.047) and pre-existing renal impairment (80.0% vs. 21.4%, p=0.017), but logistic regression did not confirm any significant independent factors (specificity 91.2%, sensitivity 61.5%, 83.0% correct predictions). Patients with nephrotoxicity received significantly fewer injections than those without (p=0.035), indicating events occurred early in treatment. Full mtDNA sequencing revealed no association between ADR occurrence and mtDNA haplogroups. All detected SNVs were sporadic, and no specific mutations, genes, or genome regions could be linked to ototoxicity or nephrotoxicity. Previously reported ototoxicity-associated mtDNA variants (m.1555A>G, m.1494C>T, m.1095T>C, m.961T>G, m.961insC(n), m.827A>G) were absent in this cohort. The m.961T>A variant was detected in one patient with ototoxicity, but its clinical significance remains unclear.", "Clinical Implications": "This study confirms that amikacin carries a higher risk of ototoxicity compared to capreomycin in MDR-TB patients, with ototoxicity occurring independently of dose or treatment duration, emphasizing the role of individual patient variability. Nephrotoxicity risk is higher in older patients and those with pre-existing renal impairment, and tends to occur early in treatment, suggesting the need for close monitoring of renal function at treatment initiation, particularly in elderly patients. The absence of previously reported ototoxicity-associated mtDNA variants in this cohort highlights the complex, multifactorial nature of ADR mechanisms and suggests that genetic screening for these specific variants may have limited utility in all populations. The high heterogeneity and low frequency of mtDNA SNVs observed underscore the challenge of identifying genetic predictors of drug toxicity. Although injectable agents are now less commonly used following the introduction of all-oral regimens, amikacin remains relevant for severe drug-resistant TB cases and other serious bacterial infections, making these findings broadly applicable."}