Pharmacotherapeutic Considerations in the Treatment of Nontuberculous Mycobacterial Infections: A Primer for Clinicians

Abstract Nontuberculous mycobacteria (NTM) can cause a variety of infections, including serious pulmonary disease. Treatment encompasses polypharmacy, with a targeted regimen of 2–5 active medications, depending on site of infection, species, and clinical characteristics. Medications may include oral, intravenous, and inhalational routes. Medication acquisition can be challenging for numerous reasons, including investigational status, limited distribution models, and insurance prior authorization. Additionally, monitoring and managing adverse reactions and drug interactions is a unique skill set. While NTM is primarily medically managed, clinicians may not be familiar with the intricacies of medication selection, procurement, and monitoring. This review offers insights into the pharmacotherapeutic considerations of this highly complex disease state, including regimen design, medication acquisition, safety monitoring, relevant drug–drug interactions, and adverse drug reactions.

Nontuberculous mycobacteria (NTM) are found naturally in the environment, especially in soil and water, with distribution varying by region, depending on climate, moisture, and other environmental conditions [1][2][3].NTM comprise ∼200 individual species of mycobacteria, excluding Mycobacterium tuberculosis and Mycobacterium leprae [1,2,4].The prevalence of NTM infection appears to be increasing, at least in part due to improved diagnostics [5,6], with Mycobacterium avium complex (MAC) representing the most common pathogen [1,6,7].Diagnosis requires a combination of clinical, microbiological, and radiographic information; isolation of NTM alone does not establish active infection [1].NTM are categorized as slow growing (eg, MAC, M. kansasii, M. marinum) or rapid growing (eg, M. chelonae, M. fortuitum, M. abscessus) based on their growth rate on culture media, taking >7 or <7 days for growth, respectively [1,[3][4][5].It is critical to distinguish species, and even subspecies, given the heterogeneous nature and implications for management [1,2,4].Chronic pulmonary disease is reported in ∼70%-90% of NTM cases, but infection involving other sites, including soft tissue and bone, is also possible [1,4,5].Disseminated disease is generally seen in the severely immunosuppressed [2].In addition, M. chimaera disseminated disease has been linked to contaminated heater-cooler units used during cardiac surgery [2].
Interdisciplinary management of NTM infection is key, given the complex pharmacologic and monitoring strategies, varied drug procurement, prolonged combination therapies, and oftentimes difficult-to-tolerate agents [8].However, given the relative infrequency of NTM infections, clinicians may feel underprepared to manage this disease.The aim of this review was to provide insights into the pharmacotherapeutics of this highly complex disease, including regimen considerations, medication acquisition, therapeutic drug monitoring (TDM), relevant drug-drug interactions, and adverse drug reactions (ADRs).Emerging therapeutic options are beyond the scope of this manuscript but have been reviewed elsewhere [9].Given global variability in the acquisition and availability of certain NTM anti-infectives, the scope of this document will be most applicable to care in the United States.
infections [2].Targeted regimens generally consist of 2-5 active medications, depending on clinical and microbiologic factors [10].However, clinical data informing treatment strategies for NTM are mostly limited to observational studies, with few randomized controlled trials available.Thus, contemporary national guideline recommendations are limited to pulmonary infections caused by select species [10].Given these limitations, treatment is often individualized based on patientspecific factors, and adherence to guideline recommendations is as low as 13% [1,3,11].
Antimicrobial susceptibility testing (AST) can be useful in designing NTM regimens, keeping important caveats in mind.First, NTM AST should be performed only for species associated with clinical disease [12,13].Second, correlation of clinical response to in vitro susceptibility patterns is established only for select species and drug pairs.For MAC isolates, clinical correlation is limited to the macrolides and amikacin (with inhaled and intravenous specific break points) [13,14].Clarithromycin is the class drug used for macrolide susceptibility testing of MAC.Azithromycin is the preferred macrolide for treatment of macrolide-susceptible MAC pulmonary disease, its susceptibility inferred from clarithromycin testing [10].The minimum inhibitory concentration (MIC) data for ethambutol, rifampin, and rifabutin have shown poor correlation with clinical response for MAC.Though these drugs are often used in the treatment of MAC, susceptibility testing is not routinely recommended because a high or a low MIC value does not determine whether these drugs are useful in treatment regimens [15].Mycobacterium kansasii treatment failure has been associated with rifampin resistance, but ethambutol (EMB) and isoniazid (INH) minimum inhibitory concentrations do not correlate well with clinical response [12,13].Susceptibility testing for other antimicrobials is recommended in the presence of macrolide resistance (for MAC) and rifampin resistance (for M. kansasii) [15].An AST panel for a slow-growing NTM may consist of amikacin, ciprofloxacin, clarithromycin, clofazimine, doxycycline, linezolid, minocycline, moxifloxacin, rifabutin, rifampin, streptomycin, and trimethoprim-sulfamethoxazole [15].
For rapidly growing mycobacteria, more comprehensive AST is performed "to guide rather than dictate therapy" [12,13,15].A wider panel may include amikacin, cefoxitin, clarithromycin, ciprofloxacin, clofazimine, doxycycline (or minocycline), imipenem, linezolid, moxifloxacin, tigecycline, tobramycin, and trimethoprim-sulfamethoxazole [15].An example susceptibility report for a rapid-growing NTM is available in Supplementary Figure 1.M. fortuitum, M. abscessus subspecies abscessus, M. abscessus subspecies bolletii, and M. smegmatis isolates should be tested for inducible macrolide resistance due to the erythromycin ribosomal methylase erm (41) gene by incubation with subinhibitory concentrations of clarithromycin for 14 days [12,[15][16][17].The erm (41) gene transfers 1-2 methyl groups to an adenine in the peptidyl region of 23S rRNA, preventing macrolide binding.The presence of either an erm (41) 274-bp deletion or erm (41) T28C point mutation independently indicates reversion to macrolide susceptibility, which is common for M. abscessus subspecies massiliense [18].Azithromycin susceptibility may be inferred from clarithromycin for rapid-growing NTM as well [12].While doxycycline or minocycline susceptibility testing may be performed for rapidly growing mycobacteria, their clinical role for the treatment of M. abscessus is questionable [19].Due to the level of expertise needed and technical challenges associated with mycobacterial AST, this is typically performed at a referral microbiology laboratory.Several microbiology laboratories can perform NTM susceptibility testing such as the National Jewish Hospital, University of Texas at Tyler Health Center, and Mayo Clinic [20][21][22][23][24][25][26][27].Waiting for AST may not be possible for disseminated and/or severe infections.Thus, empiric therapy based on the available information on mycobacterial organism susceptibility patterns is recommended until AST results are available.
Outpatient parenteral antimicrobial therapy (OPAT) is a key component in the multimodal management of NTM infections, particularly those with cavitary MAC lung disease and those with multidrug-resistant organisms such as M. abscessus [10,28].Infectious diseases-trained clinical pharmacists are well positioned to assist prescribers in the management of patients requiring OPAT for NTM, and the need for their services has expanded in the United States [29,30].Prospective serial laboratory and safety monitoring is of paramount importance, as duration-dependent toxicities are well described with several agents used for NTM treatment [28].Recommended monitoring parameters are summarized in Table 1.In many instances, data supporting the frequency of laboratory monitoring are limited to low-quality evidence or expert opinion [28].Thus, monitoring schedules should be individualized based on patient-specific factors such as severity of infection, duration of therapy, and presence of baseline organ dysfunction and/ or drug interactions.
Although intravenous antibiotics may be unavoidable in some scenarios, vascular access device placement comes with associated risk and significant health care resources.Emerging oral therapies with broad-spectrum activity against rapid-growing mycobacteria, such as oxazolidinones and omadacycline, may provide an alternative to historically intravenous regimens in some patients [9].Clinical data supporting these agents are limited, although ongoing clinical trials may clarify their role in therapy [9,[41][42][43][44][45][46].Nonetheless, there remains an urgent need for additional, well-tolerated oral options with activity against species with limited susceptibility.

Medication Acquisition
Clinicians are often tasked with procuring NTM medications, which can involve complex acquisition models.While Table 2 summarizes the general distribution models, a few unique scenarios are further described in this section.
With the exception of clofazimine, NTM medications are subject to prescription insurance coverage.As many are high cost, have niche indications, or are used off-label for NTM treatment, they may require prior authorization (PA) and/or appeals to be covered.Typically, the information required includes diagnosis (or International Classification of Diseases, 10th Revision, code), identified organism, and reasons why alternative therapies are not suitable.This may include allergies, drug interactions, or previous untoward effects.Susceptibility information can be important, especially if no other susceptible options are available.Referencing primary literature or national guidelines can improve PA approval rates, but some may still require a peer-to-peer discussion regarding the clinical case and need for a specific medication.Building electronic health record (EHR) templates may be helpful.A sample appeal letter can be found in Supplementary Figure 2.
Even with "covered" medications, patients may still face exorbitant copays.The drug manufacturer may have financial assistance programs or "copay cards" available.Additionally, philanthropic or research organizations may offer financial support.Two such organizations are The Assistance Fund [53] and HealthWell Foundation [54].Programs vary based on what federal poverty level equivalent is eligible for assistance.Notably, Medicare recipients and other government insurances may be excluded from assistance.Institutions may have internal financial assistance programs, with more leniency in terms of federal poverty level.Programs can have different enrollment periods, limitations on financial aid provided, and annual capacities.
Inhaled liposomal amikacin is currently available via 2 specialty pharmacies.A combined prescription order and patient assistance form is available online [47], which can facilitate financial assistance and first-delivery in-home nurse teaching with the included inhalation device.The patient's signature is required to enroll in the teaching program.Insurance benefits investigation and PA can be completed by internal clinic staff or by the manufacturer via the enrollment form [48].
Bedaquiline is currently only available through 1 specialty pharmacy via a special ordering form that is faxed to the distributor.As the loading and maintenance dosing are vastly different, clinics can prepopulate a blank form and save for future use [49].Once the copay is collected, medication is delivered to a medical clinic or office, and the patient should be contacted for pickup.The medication cannot be delivered to the patient's home.
Clofazimine is no longer commercially available in the United States but is available via expanded access under investigational drug access.If only prescribing for 1 patient, a Single Patient Investigational New Drug (SPIND) application can be submitted to the Food and Drug Administration (FDA) [50].However, if multiple patient treatments are anticipated, clinicians are asked to initiate a multiple patient program at their institution [51].Both avenues require coordination with the pharmaceutical company and the local institutional review board (IRB).Once approved, patient consent is obtained, and patients are individually enrolled via the manufacturer's online portal [52].A patient-specific 90-day drug supply is requested through the portal by an ordering clinician listed on the IRB protocol and shipped to a research pharmacy.Clofazimine is provided free of charge (no insurance or copayment required), and patients pick up the medication from the clinic or research pharmacy.Since the coronavirus disease 2019 pandemic, some states may allow research pharmacies to mail the medication to the patient.
Medications with limited distribution models and/or dispensing via specialty pharmacies may not appear in third-party refill records.Tasking a team member with maintaining a tracking document to ensure timely refills and address insurance issues may facilitate improved oversight and adherence to therapy.

Therapeutic Drug Monitoring
Analogous to other infectious diseases, TDM in the treatment of NTM is valuable when relationships between exposures and efficacy or toxicity have been established, therapies demonstrate pharmacokinetic variability, patient organ function changes, patients take interacting medications, barriers to adherence exist, disease prognosis is guarded, or immunologic defects are persistent and/or severe [55].
A key driver of TDM is the establishment of exposureresponse relationships with efficacy or toxicity.Pharmacokinetic/ pharmacodynamic (PK/PD) efficacy targets are frequently established through in vitro data and infection models and/or retrospective analyses of clinical data [55].While much progress has been made in recent years using in vitro models such as the hollow-fiber model, retrospective and prospective clinical validation remains limited for most NTM therapies.
Relationships between exposures and associated toxicities play a particularly important role in NTM treatment due to the extended duration of therapy required, variable tolerability, occasional overlapping toxicities, and limited therapeutic options.Furthermore, physical characteristics of body composition, such as cachexia, may impact drug pharmacokinetics [56].Thus, TDM may be used to minimize the impact of toxicities on regimen selection, adherence, and patient outcomes.
In clinical practice, routine TDM is standard of care for patients receiving intravenous amikacin for NTM given the risk of serious adverse events, notably nephrotoxicity and ototoxicity.A calculated maximum concentration (C max ) is recommended, ideally extrapolated using serum drug levels collected at 2 and 6 hours following administration to avoid sampling during the redistribution phase.The target C max for intravenous amikacin is 35-45 mcg/mL (15-mg/kg dose) or 65-80 mcg/mL (25-mg/ kg dose), while the target trough should be undetectable [15].TDM for additional antimycobacterials should be considered in clinical scenarios with concern for low drug exposure, including patients with inadequate clinical or microbiological response [10].Drug exposure may be compromised in the setting of malabsorption (bariatric surgery, severe gastrointestinal disease) and drug interactions, which could lead to treatment failure and/or development of resistance [15].A comprehensive review of the data supporting the PK/PD targets of antimycobacterials was recently compiled [57], and C max targets for select oral agents are included in Table 1 [31].
Implementation of TDM in the clinic is complicated for several reasons.First, very few US laboratories possess clinically validated antimycobacterial drug assays.Thus, a reference laboratory must be identified and workflows established.This requires substantial coordination between the multidisciplinary team, the facility's laboratory, and the external laboratory to ensure that TDM is successfully conducted.Once a reliable clinical assay is available, the TDM approach and interpretation add another layer of complexity.Although antimycobacterial TDM has historically focused primarily on the determination of peak or trough concentrations, growing evidence suggests that population PK and Bayesian dosing adjustments may play a greater role in optimizing antimycobacterial regimens, similar to tuberculosis [57,58].While these approaches have the potential to give patients the best chance at clinical success through individualized drug therapy, use of the required software, result interpretation, and dose adjustments require additional resources and expertise.Despite logistical challenges, antimycobacterial TDM using a reference laboratory for testing has been clinically implemented in at least 1 large academic medical center in the United States [59].Referral laboratories for antimycobacterial TDM include University of Florida-Shands and National Jewish, and antimycobacterial TDM is in exploration at Mayo Clinic [20][21][22][23][24][25]38].

Adverse Drug Reactions
Managing NTM infections often involves managing ADRs.Common and notable ADRs with mitigation strategies, management recommendations, and additional considerations are listed in Table 3.
The most frequently encountered ADRs involve the gastrointestinal (GI) tract and may occur with any NTM-active therapy.Effects include diarrhea, nausea, vomiting, abdominal pain, and/ or dysgeusia.Hepatotoxicity has been associated most notably with rifamycins, bedaquiline, tetracyclines, imipenem-cilastatin, and macrolides.While not evaluated systematically, the FDA  provides guidance on therapy discontinuation when liver injury occurs in clinical trials (Table 3 legend) [61].Nephrotoxicity has been causally linked to trimethoprim-sulfamethoxazole and is well documented with aminoglycoside use.Trimethoprim inhibits cellular transport proteins in the proximal tubule, responsible for serum creatinine secretion, leading to a mild pseudo-elevation of creatinine within hours of medication initiation [75].However, trimethoprim-sulfamethoxazole can also rarely lead to acute interstitial nephritis or acute tubular necrosis.Ototoxicity (primarily cochlear toxicity) with long-term intravenous or inhaled amikacin is primarily irreversible [80]; however, hearing loss or tinnitus is generally reversible within a month of treatment discontinuation with macrolide therapy [81].Ethambutol or oxazolidinones may cause optic neuritis and/or peripheral neuropathy, which tend to be durationdependent, generally occurring after at least 1 month of therapy [82].Oxazolidinones can also result in myelosuppression with prolonged use, generally thrombocytopenia and/or anemia, more so with linezolid than tedizolid [39,83].Trimethoprimsulfamethoxazole, rifamycins, and β-lactams have also been associated with hematologic toxicity [9,84].Treatment modification should be considered with rapid decreases in hemoglobin/hematocrit, neutrophils, or platelets while on therapy in those without an alternative etiology for myelosuppression.
Idiosyncratic dermatologic and cosmetic ADRs may occur.Rifamycins are associated with a benign orange discoloration of body fluids, including sweat, saliva, urine, and tears [9].Similarly, clofazimine can cause an orange-brown discoloration of body fluids and, much more commonly, the skin [35].Tetracyclines also carry photosensitivity precautions.Blue-gray skin discoloration may also occur with long-term use, particularly with minocycline.Given the polypharmacy nature of treating NTM, patients are often at increased risk of ADRs when multiple agents with similar risks of toxicity are used in combination.Some centers may use a phased-in approach (eg, stagger medication starts on a weekly basis) to assess for adverse reactions and aim to improve tolerability.Additionally, lower doses may be used for a week or 2 before progressing to the full dose.Clinicians are encouraged to report ADRs to the FDA Adverse Event Reporting System (FAERS) via MedWatch to aid in postmarketing surveillance efforts [85].overlapping toxicity, or (3) diminished oral absorption due to chelation by concomitantly administered multivalent cationcontaining compounds, such as calcium, magnesium, iron, aluminum, and zinc.In addition, elimination of vitamin Kproducing bacteria by antimicrobials may contribute to alterations in coagulation for patients taking warfarin.Comprehensive references and interaction databases should be consulted when providing patient care [4]; Table 4 provides a nonexhaustive summary of relevant drug interactions.Given the potential for discrepancies and variable performance in interaction identification and severity classification among databases, expert review by a clinical pharmacist can aid in the interpretation, mitigation, and management of drug interactions [86].
Interaction potential can be influenced by dosage, duration, inhibition or induction potency, and metabolic enzyme binding affinity.Most metabolic interactions are manifested through the CYP system of enzymes.CYP inducers can increase metabolism of CYP substrates (resulting in decreased substrate levels), while CYP inhibitors can decrease metabolism of CYP substrates (resulting in increased substrate levels).Thousands of medications undergo some degree of metabolism by the CYP enzymes.CYP3A4/5 is the most predominant in humans and is responsible for many interactions.Medications subject to CYP-based interactions are maintained on the FDA's website [100].
An additional type of metabolic interaction involves the use of linezolid with serotonergic agents.Linezolid is a weak and reversible inhibitor of monoamine oxidase (MAO).This inhibition blocks oxidative deamination to cause the accumulation of endogenous catecholamines (serotonin and norepinephrine).Combining linezolid with serotonergic or adrenergic agents, or tyramine-containing foods, can lead to increased blood pressure or the rare occurrence of serotonin syndrome [94,95].Of note, contemporary data suggest an incidence of linezolid-associated serotonin syndrome well below 1%, even when co-administered alongside other agents with serotonergic activity [94].Tedizolid is a newer oral oxazolidinone believed to have an even lower risk of these interactions due to its weaker MAO inhibition and lower central nervous system penetration [97].
Organ toxicity may also occur.QTc prolongation can increase the risk of ventricular arrhythmias.Many medications can prolong QTc, and the risk of arrhythmia can be increased by combining these agents.Among NTM medications, fluoroquinolones, macrolides, bedaquiline, and clofazimine have all been reported to prolong QTc intervals.Electrocardiography and electrolyte monitoring are recommended, particularly when these agents are used in combination, and close repletion of potassium and magnesium may be indicated to mitigate toxicity [62].Combining NTM agents with hepatotoxic potential may increase the risk of an adverse event [101].Likewise, patients could experience higher nephrotoxicity rates when multiple nephrotoxic agents are combined [102].
Additional miscellaneous interactions exist.Trimethoprim has been associated with an antikaliuretic effect, which can contribute to a significant hyperkalemic response when administered with potassium-sparing agents like angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, or spironolactone [98,99].Fluoroquinolones can cause hypoglycemia by increasing insulin release via blockade of adenosine triphosphatesensitive potassium channels in pancreatic β cells [90].This effect can be accentuated in patients taking oral hypoglycemic therapy.Orally administered fluoroquinolones and tetracycline analogs can have their absorption diminished via chelation, when concomitantly administered with multivalent cations.Patients should be advised to separate administration of these agents from any multivalent cation to ensure adequate absorption; specific time intervals for separation are listed in Table 4 [ 63-65, 69, 76, 77, 91, 103].Trimethoprim-sulfamethoxazole interacts with methotrexate through anion transport inhibition, which leads to an additive inhibition of dihydrofolate reductase [98].This interaction can contribute to pancytopenia.

Multidisciplinary NTM Care Team
A multidisciplinary team approach is ideal in the management of NTM infection.It may be said that NTM care "takes a village."The NTM care team encompasses several disciplines, including but not limited to pulmonology, infectious diseases, radiology, microbiology, respiratory therapy, primary care, and pathology (Figure 1) [104].Diagnosis of NTM disease is complex, at times taking as long as 20 months, and requires communication and coordination among pulmonologists, infectious disease specialists, radiologists, and clinical microbiologists [105].Disease management may additionally involve respiratory therapists, surgeons, nurses, and/or dieticians [106].Clinical pharmacists specializing in NTM can assist with antimycobacterial dosing and TDM, ADR management, and drug interaction mitigation; guidance on implementing a pharmacist resource for NTM is available in Supplementary Table 1 [59].Social workers and patient support groups can be incorporated to help patients cope with psychosocial and financial concerns [106].Clinical outcomes research with the multidisciplinary team may be further supportive of this approach.

CONCLUSIONS
NTM infections are increasing in prevalence, and their management involves highly complex treatment and monitoring considerations.Effective treatment of NTM requires polypharmacy, often based on interpretation of unvalidated susceptibility breakpoints.Emerging evidence suggests a role for TDM in this patient population, and intimate knowledge is required of drug interactions, adverse effect management, pharmaceutical acquisition, and financial assistance.A multidisciplinary care team is ideal for the care of NTM patients, with a growing role for clinical pharmacist involvement.

Table 1 . Recommended Safety Monitoring Schedule for Commonly Used NTM Therapies
Pharmacotherapeutic Considerations for NTM Treatment • OFID • 3

Table 2 . Outpatient NTM Medication Acquisition
Abbreviations: FDA, Food and Drug Administration; ITC, infusion therapy center; IV, intravenous; NTM, nontuberculous mycobacteria.a Rifabutin has an FDA indication for the prevention of disseminated M. avium in patients with HIV.