Colonization with Nontuberculous Mycobacteria Is Associated with Positive Tuberculin Skin Test Reactions in the Common Marmoset (Callithrix jacchus)

Abstract

Mycobacterium tuberculosis infections can result in significant morbidity and mortality in nonhuman primate colonies. Preventative health programs designed to detect infection routinely include tuberculin skin testing (TST). Because Mammalian Old Tuberculin used for TST contains antigens common to a variety of mycobacterial species, false-positive results can occur in animals sensitized to nontuberculous mycobacteria (NTM). Over 11 mo, a large colony of common marmosets (Callithrix jacchus) demonstrated a 3.6% prevalence of equivocal or positive TST reactions (termed ‘suspect reactions’). Culture of gastric aspirates, bronchoalveolar lavage fluid, and feces revealed a single animal with a positive fecal culture for Mycobacterium gordonae. PCR amplification of M. gordonae DNA in feces collected from animals with suspect TST reactions (demonstrating a 66.7% colonization rate) and colony controls (demonstrating a 14.3% colonization rate) revealed a significant association between suspect TST reactions and intestinal colonization. Gross and histopathologic evaluation revealed a multifocal lymphadenopathy and granulomatous lymphadenitis in 2 of 4 TST-positive marmosets examined. Counter to expectations, granulomatous lymphoid tissue was culture-positive for M. kansasii rather than M. gordonae. Detection of M. gordonae in the feces of TST-suspect animals likely represents an apathogenic intestinal colonization that may serve as an indicator of NTM exposure, whereas evidence of histopathologic disease is associated with the more pathogenic M. kansasii. Although a high index of suspicion for M. tuberculosis should always be maintained, colonization with NTM organisms represents a cause of suspect TST reactions in common marmosets.

Tuberculosis infections pose a serious risk to nonhuman primate colonies and hold the potential for significant contributions to morbidity and mortality. In addition to the compromise of research efforts, tuberculosis infections represent a zoonotic risk to human handlers. Because of these concerns, research facilities housing nonhuman primates implement preventative health care programs that routinely include a variety of screening methods to detect tuberculosis infections. Despite limitations, the mainstay among the screening methods used for diagnosis of M. tuberculosis infections remains the tuberculin skin test (TST).⁸

The TST response is based on the development of a delayed hypersensitivity (type IV) immune response after intradermal injection of a specified amount of ‘Mammalian Old Tuberculin’ into the superior palpebra, where the result is monitored easily. Erythema and swelling are monitored daily between 24 and 72 h and graded on a scale of 0 through 5.^{3,19, 25, 37} Grades 4 and 5 consist of significant swelling with ptosis or necrosis and are considered positive. Grade 3, which consists of minimal swelling with or without erythema, is considered an equivocal result. In addition to the subjective nature of the TST, the test is prone to false-negative and false-positive results.^28,29 False-negative results have been documented to occur in latently infected animals, anergic animals with advanced disease, and animals with defects in cellular immune responses.^7,9,10,15,36 Because Mammalian Old Tuberculin may contain antigens common to a variety of mycobacterial species, false-positive results can occur in animals sensitized to nontuberculous mycobacterial (NTM) species.^17,21 Additional causes of false-positive results include traumatic injection, prior exposure to Freund complete adjuvant, and vaccination with the Bacillus Calmette–Guerin vaccine.^29,34,42

Interpretation of the TST result in part depends on a risk assessment for the population of animals subject to screening. The veterinary response to an equivocal (grade 3) or higher reaction may be tempered in light of the animal's source, clinical history, and colony history but generally involves room quarantine and implementation of additional personal protective equipment requirements, pending results of further diagnostic assays. Additional diagnostic measures may include radiographs, culture of bronchial or gastric aspirates, in vitro assays of the cellular immune response to mycobacterial antigens, survey for antibody response to mycobacterial antigens, and euthanasia for gross necropsy and histopathologic assessment.^6,16,23

The index of suspicion for tuberculosis infections also can vary depending on the species of nonhuman primate under consideration. M. tuberculosis is a member of a complex of closely related organisms known as the M. tuberculosis complex (MTC), which also includes M. bovis, M. africanum, M. canetti, and M. microti.⁴⁶ Old World primates are considered to be most susceptible to infections with MTC organisms, whereas New World primates are thought to be relatively resistant to naturally acquired infection.^26,33 Although sporadic infection with MTC organisms in New World primates has been noted, these reports often document cohousing with macaque species.^1,18,27,32 Colonization with NTM appears to be more common in New World species and includes asymptomatic infections with M. avium complex species, M. kansasii, and M. gordonae.^1,5,43

Here we describe an investigation of the cause of equivocal and positive TST reactions in a colony of common marmosets and report an association with exposure to NTM. A description of the clinical work-up performed to exclude presence of MTC organisms is included.

Materials and Methods

Animals.

Common marmosets (Callithrix jacchus) were housed at the New England Primate Research Center (Southborough, MA) and maintained in accordance with the Guide for Care and Use of Laboratory Animals.²⁰ The facility is AAALAC-accredited. Colony animals are maintained under an animal holding and breeding protocol approved by Harvard Medical School's Standing Committee on Animals. Animals are housed in stainless steel caging in pairs or groups of 7 to 8 with room temperature maintained at 78 ± 4 °F (26 ± 2 °C) and humidity between 30% and 70%. Colony animals received commercial marmoset chow (Harlan Teklad New World Primate Chow 8791, Madison, WI) supplemented with a combination of fresh fruits, vegetables (JW Lopes, Chelsea, MA), eggs (Sysco Boston, Norton, MA), and a seed and nut mix (Monkey Jumble, PMI Nutrition International, Brentwood, MO). Depending on the housing status of the animals, ad libitum water was delivered by polycarbonate water bottles filled directly from the tap or by an automatic water-delivery system. The water supply is municipal, and microbial quality assurance for coliforms, Giardia, and cryptosporidium is performed by the municipality. Environmental enrichment consists of forage, toys, nest boxes, and music. Two marmosets (nos. 6 and 9) were assigned to a dietary study and received an alternative chow (Mazuri Callitrichid Diet, 5MI5, PMI Nutrition) in addition to the supplements detailed earlier. A third marmoset (no. 8) was assigned to a study examining learning and cognition. The remaining animals were not assigned to research studies.

Preventative health care and clinical diagnostics.

Preventative health care for the marmoset colony is performed triannually. Animals were sedated with ketamine HCl (10 to 15 mg/kg IM, Fort Dodge Animal Health, Fort Dodge, IA) for complete physical examination, phlebotomy for serum banking, and TST. A volume of 0.05 mL of Mammalian Old Tuberculin (Lot 423X for the April 2008 TST and Lot 425X for the August 2008 and February 2009 TST; Synbiotics Corporation, San Diego, CA) was injected intradermally in the superior palpebra by using a 29-gauge needle. The test was read during cageside observation by members of the veterinary staff at 24, 48, and 72 h after inoculation and graded on a scale of 1 through 5. TST results from August 2008 and February 2009 were read by 3 independent observers. Reactions of grade 4 to 5 are considered positive, with a grade 3 representing an indeterminate result. For the purposes of this investigation, marmosets demonstrating a TST reaction of grade 3 and higher were considered suspects for mycobacterial disease (termed ‘TST suspect animals’) and underwent a clinical workup. Diagnostic assays performed included phlebotomy for complete CBC, thoracic radiography, and culture of gastric aspirates, bronchoalveolar lavage fluid, and fecal material. Phlebotomy was performed according to standard techniques with blood collected from the femoral vein. Gastric and bronchoalveolar aspirates were performed under ketamine sedation. Sterile saline was flushed and immediately withdrawn through a sterile 8-French pediatric feeding tube inserted into the esophagus or trachea. Fresh fecal samples were collected from clean paper cage pan liners and stored in sterile cryovials at −80 °C. Mycobacterial cultures were incubated for at least 60 d (Massachusetts Department of Public Health, Division of Tuberculosis Prevention and Control, Mycobacteriology Laboratory).

Fecal PCR.

DNA was isolated and purified from marmoset fecal material (QIAamp DNA Stool Mini Kit, Qiagen, Valencia, CA) according to the manufacturer's instructions. PCR amplification of M. gordonae DNA was performed by using a nested PCR reaction and primers directed against the 85A antigen sequence.¹² Primers G1 and G3 were used to generate the outer PCR product (407 bp), and primers G2 and G3 were used to generate the inner PCR product (144 bp). Boiled suspensions of a M. gordonae mycobacterial colony (strain 14470, ATCC, Manassas, VA) were used as the positive PCR control, and sterile distilled water was used as the negative PCR control. All reactions were adjusted to 50 µL and contained 10× reaction buffer with 15 mM MgCl₂ (Roche Applied Science, Indianapolis, IN), 1 µL dNTP, 0.9 to 1.1 µL of each primer, and 0.5 U Taq DNA polymerase. The PCR thermocycling conditions were: 94 °C for 2 min followed by 35 cycles of 94 °C for 30 s, 69 °C for 45 s, and 72 °C for 60 s. A final extension step was performed at 72 °C for 7 min. Products were separated on a 2% agarose gel containing ethidium bromide and visualized under UV light. The 144-bp amplification products of a subset of samples were cloned (Topo TA Cloning Kit, Invitrogen, Carlsbad, CA) and sequenced (Retrogen, San Diego, CA) from the G2 and G3 primers to verify that product sequences corresponded to that of M. gordonae. The program BLASTN (http://www.ncbi.nlm.nih.gov) was used to align sequences against those in various databases.

Pathology.

Of the 5 animals demonstrating overt positive TST reactions of grade 4, 4 marmosets were euthanized by intravenous overdose of pentobarbital (in excess of 50 mg/kg) for complete necropsy and histopathologic examination. The remaining marmoset (no. 8) was assigned to an ongoing study and was not euthanized. Necropsy was performed within 30 min of death, and representative sections of all major organs were collected, fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 5 µm, and stained by using hematoxylin and eosin. Ziehl–Neelsen and Fite–Faraco stains were performed on tissues containing suspect lesions with granulomatous inflammation.

Data analysis.

Analysis was performed by using Stata software (Stata Press, College Station, TX). Statistical comparisons were performed by using the Fisher exact test. A P value of less than or equal to 0.05 was considered significant.

Case Study

Results of tuberculin skin testing.

After routine preventative health care in April 2008, 5 marmosets were noted to have suspect TST reactions of grade 3 or 4 (Table 1, Figure 1). These animals included 3 male and 2 female marmosets ranging in age from 1 to 5 y that were housed in group cages with 7 or 8 animals per cage. Marmosets nos. 3 and 4 were cohoused; the other 6 animals housed within this cage remained TST negative. With the exception of marmoset no. 3 which was found to have thin body condition, physical examinations were within normal limits. All 5 marmosets underwent additional diagnostic testing, including thoracic radiographs, mycobacterial culture of gastric lavage fluid, and mycobacterial culture of feces. Mycobacterial culture of bronchoalveolar lavage fluid was performed on a subset of animals (Table 1). Thoracic radiographs were unremarkable. Because of the colony's closed status, lack of opportunity for exposure to macaque species, and a rigorous employee occupational health program, there was not a high index of suspicion for M. tuberculosis infection. Pending mycobacterial culture results, a ban on movement of animals between the marmoset housing rooms was instituted. Results of the mycobacterial cultures were negative except for that of marmoset no. 4, which demonstrated growth of M. gordonae on fecal culture; this isolate was speciated by using a gene probe. Neither M. tuberculosis nor M. bovis was detected in any of the cultures performed.

Table 1.

Results of tuberculin skin testing (TST) and additional diagnostic assays for 10 marmosets suspected of mycobacterial disease

	April 2008 TSTa			August 2008 TSTa			February 2009 TSTa			Additional diagnosticsb
	24 h	48 h	72 h	24 h	48 h	72 h	24 h	48 h	72 h	Rads	GL	BL	FC	Pathology	TC
1	3	3	2	0	0	0	0	0	0	—	—	—	—
2	0	0	3	3	3	3	0	0	0	—	—	—	—
3	0	0	4	4	4	4	0	0	0	—	—	—	Granulomatous disease	+
4	0	0	3	0	3	3	3	2	3	—	—		+
5	3	4	4	3	4	4	4	3	3	—	—	—	No findings	—
6				0	0	3	3	2	2	—			—
7							3	3	2				—
8							3	3	4				—
9							0	4	4	—			—	No findings	—
10							0	4	4	+			—	Granulomatous disease	+

^aTST graded on scale of 0 to 5, with grades 3 and above indicating suspicion of mycobacterial disease. Blank tabular cells indicate TST reactions of grade 2 or lower.

^bAdditional diagnostics performed include thoracic radiographs (Rads), mycobacterial culture of gastric lavage fluid (GL) or bronchoalveolar lavage fluid (BL), mycobacterial culture of feces (FC), necropsy (Pathology), and mycobacterial culture of tissues (TC). A blank cell indicates that the diagnostic test was not performed for that marmoset.

Figure 1.

Representative tuberculin skin test (TST) reactions in common marmosets. (A) Depicts a grade 3 TST reaction. (B) Depicts a grade 4 reaction.

TST was repeated in August 2008. Marmosets nos. 2 through 5 continued to demonstrate suspect TST reactions, and an additional suspect animal was identified (Table 1). Marmoset no. 6 was a pair-housed, 12-y-old male animal in good physical condition. Marmoset no. 1 was TST-negative at this time, with a grade 0 reaction noted on all readings. Fecal DNA isolated from marmosets nos. 2 through 6 and 5 colony controls (all with negative TST reactions) was assayed for the presence of M. gordonae DNA (Figure 2). Of the animals assayed, none of the colony controls (0%) demonstrated presence of M. gordonae DNA in feces, whereas M. gordonae DNA was isolated in feces from 3 of the 5 animals with suspect TST reactions (60.0%; Fisher exact P value = 0.08). Sequencing of 144-bp amplification products from marmosets nos. 2 and 3 and the positive-control ATCC strain revealed a 97.8% average identity with the reported sequence of the gene encoding the M. gordonae 85A antigen.

Figure 2.

PCR detection of M. gordonae DNA from marmoset feces. Lane 1, molecular weight marker; lanes 2 through 6, colony control animals with negative tuberculin skin test reactions; lanes 7 through 11, TST suspect animals (lane 7, marmoset no. 4; lane 8, marmoset no. 6; lane 9, marmoset no. 2; lane 10, marmoset no. 3; and lane 11, marmoset no. 5); lane 12, positive control ATCC M. gordonae strain 14470. Lanes 8 through 10 and 12 demonstrate a 144-bp PCR product consistent with presence of M. gordonae DNA.

An additional 4 marmosets with suspect TST reactions were identified in February 2009. These animals (2 male, 2 female) ranged in age from 2 to 9 y, were pair-housed, and were in good physical condition (Table 1). Thoracic radiographs of marmoset no. 10 demonstrated enlarged tracheobronchial lymph nodes. Marmosets nos. 1, 2, and 3 were TST-negative during this round of preventative health care. The overall total of 10 marmosets with suspect TST reactions represented a colony prevalence of 3.6%. With the exception of marmosets nos. 3 and 4 which were cohoused, as previously described, all cagemates of TST suspect animals remained TST negative. There was no discernable pattern among marmosets demonstrating suspect TST reactions; animals were distributed throughout the 4 marmoset housing rooms, were housed either in pairs or groups, and had water supplied either by automatic water delivery systems or polycarbonate bottles.

Survey for M. gordonae DNA in marmoset feces.

A more extensive examination for the presence of M. gordonae DNA in marmoset feces was performed in an effort to examine the prevalence and epizootiology of infection. Fresh feces were collected daily over 3 consecutive days from TST suspect animals, cohoused control animals demonstrating negative TST reactions, and separately housed control animals demonstrating negative TST reactions (Table 2). Presence of M. gordonae DNA was confirmed in 6 of 9 animals with suspect TST reactions (66.7%) compared with 2 of 14 control animals (14.3%; Fisher exact P value = 0.02), confirming a statistically significant association between suspect TST reactions and intestinal colonization with M. gordonae. Shedding of the organism appeared to be intermittent; not all of the fecal samples from the TST suspect animals demonstrated presence of M. gordonae on all of the days that fecal collection was performed.

Table 2.

PCR detection of Mycobacterium gordonae DNA isolated from marmoset fecal samples collected over 3 consecutive days

	TST suspect animals			Cagemates of TST suspect animals			Colony animals
Marmoset no.	Day 1	Day 2	Day 3	Day 1	Day 2	Day 3	Day 1	Day 2	Day 3
1	NE	NE	NE	NE	NE	NE	—	—	—
2	—	—	—	—	—	—	—	—	—
3	+	+	+	NE	NE	NE	—	—	—
4	—	+	—	—	—	—	—	—	—
5	—	+	+	NE	NE	NE	—	—	—
6	—	—	—	—	—	—	—	+	—
7	+	+	—	—	—	—	—	—	—
8	+	+	+	+	—	—	NE	NE	NE
9	—	—	—	—	—	—	NE	NE	NE
10	—	—	+	—	—	—	NE	NE	NE

+, amplification of M. gordonae DNA; —,lack of amplification of M. gordonae DNA; NE, not evaluated

Gross necropsy and histopathology.

Four animals demonstrating overt positive TST reactions of grade 4 were euthanized for complete necropsy and histopathologic evaluation. Animals selected for necropsy included a prior TST responder with poor body condition (marmoset no. 3), a recurrent TST responder (marmoset no. 5), and 2 recently identified TST responders (marmosets nos. 9 and 10). The necropsy of marmoset no. 3 revealed the most extensive gross lesions, with marked lymphadenomegaly of multiple mesenteric, tracheobronchial, axillary, and inguinal lymph nodes. Histologically there was marked granulomatous lymphadenitis involving all examined lymph nodes (Figure 3) In addition there was multifocal granulomatous hepatitis, granulomatous splenitis, and granulomatous osteomyelitis of the left femur. Marmoset no. 10 had mild lymphadenomegaly of the tracheobronchial lymph nodes, which histologically corresponded to multifocal granulomatous lymphadenitis. Marmosets nos. 5 and 9 lacked gross or histologic lesions consistent with mycobacterial disease in all tissues examined. Comorbid findings in these animals included: lymphoplasmacytic nephritis (2 of 4), vacuolar hepatopathy (4 of 4), and a focal bone callus (1 of 4). Ziehl–Neelsen and Fite–Faraco staining of sections of lymph nodes, liver, spleen, and bone marrow failed to reveal acid-fast organisms. Mesenteric (marmosets nos. 3, 5, and 9) and tracheobronchial (marmoset no. 10) lymph nodes collected in a sterile manner at necropsy and frozen at −80 °C were submitted for mycobacterial cultures. Samples from marmosets demonstrating lesions (nos. 3 and 10) were culture-positive for M. kansasii, whereas those from marmosets lacking lesions (nos. 5 and 9) were culture-negative. Speciation of these isolates was determined based on biochemical assays, growth characteristics, and typical colony morphology. PCR amplification of the gene encoding for the mycobacterial 65-kDa heat shock protein (hsp65) was performed on the culture isolate from marmoset no. 3.⁴⁴ Direct sequencing of the 439-bp amplification product revealed a 99% identity with the reported hsp65 gene sequence for M. kansasii. The isolate from marmoset no. 10 was not available.

Figure 3.

Marked lymphadenomegaly of (A) inguinal and (B) mesenteric lymph nodes in marmoset no. 3. (C) Mesenteric lymph node with roughly 30% of the architecture effaced by sheets and small aggregates of macrophages and neutrophils. Magnification, ×40. (D) Mesenteric lymph node illustrating a typical granuloma (region indicated by arrow in panel C] with a central region of necrosis surrounded by lymphocytes and epitheliod macrophages with fewer neutrophils. Magnification, ×400.

Discussion

Here we report a significant association between fecal PCR detection of M. gordonae and the observation of suspect TST reactions in common marmosets. TST positivity was associated with multifocal granulomatous disease affecting a variety of tissues in a subset of the colonized marmosets. Counter to expectations, granulomatous lesions were culture-positive for M. kansasii rather than M. gordonae. Results suggest that NTM are responsible for the observed suspect TST findings in this group of animals, although the relative contribution of each mycobacterial species (M. kansasii compared with M. gordonae) to the suspect TST reactions remains in question. Although PCR detection of M. gordonae was associated with TST reactivity, occult M. kansasii colonization may have contributed as well. The presence of histopathologic lesions was associated with M. kansasii colonization.

Mycobacterial disease has not been examined thoroughly in New World species of primates. The scientific literature documenting MTC infections predominantly consists of isolated case reports that often describe cohousing of New World primates with macaque species.^18,27,32 A survey of 68 New World primates demonstrated a 7% prevalence of M. tuberculosis detected by PCR assay for the Mtp40 gene.¹ Because the zoological collection in the cited report¹ was located in a region endemic for human tuberculosis, the animals may have acquired the infection through contact with tuberculous people. Although occult MTC infection cannot be definitively excluded as the cause of the suspect TST reactions observed in our closed colony of marmosets, the repeated inability to detect MTC organisms by culture and lack of a source for exposure considerably reduces the index of suspicion.

In contrast to the low incidence of naturally acquired MTC infections, colonization with NTM appears to be more common in New World species of primates. The same previously cited study involving a zoological collection of New World primates¹ demonstrated greater than 57% prevalence of NTM colonization. M. chelonae, M. intracellulare, M. nonchromogenicum, and M. fortuitum represented the most commonly detected isolates.¹ Clinical disease was not associated with colonization by these organisms.¹ Other reports have described granulomatous lymphadenitis and granulomatous pneumonia in New World species of primates associated with M. kansasii, M. abscessus, and M. asiaticum.^5,24,41

Person-to-person transmission of NTM organisms is not thought to occur.^13,45 This theory is supported by our negligible detection of M. gordonae in the feces of TST-negative marmosets cohoused with animals demonstrating suspect TST reactions and shedding M. gordonae. Exposure to NTM is secondary to the ubiquitous presence of these bacteria within the environment. NTM are normal inhabitants of soil, water, and dust. Water distribution systems represent the most important environmental source of NTM.¹⁴ The hydrophobic nature of the cell wall and subsequent incorporation into biofilms, resistance to disinfectants such as chlorine, and ability to grow in a wide range of pH levels allow for NTM to persist in both institutional and home plumbing systems.^13,39 Potable water supplied through a recirculating-type water-heating system can be an important source of M. avium infection as demonstrated in SIV-inoculated rhesus macaques from our colony.³⁰ For these reasons, the water distribution system is the most likely source of NTM exposure in this case series. A planned avenue of research is the systematic survey of the watering systems to characterize the population of NTM present and confirm the source of exposure to M. gordonae and M. kansasii.

In humans, systemic immune deficits and impaired local defenses such as chronic obstructive pulmonary disease, cystic fibrosis, and gastroesophageal reflux disease predispose to NTM-associated infection. The majority of cases present as pulmonary disease. In addition, lesions may be detected within lymph nodes, skin, soft tissue, or bone.³⁵ NTM colonization without human disease or pathology is also common. Colonization may be difficult to distinguish from infection that is associated with underlying disease. Symptomatology, radiographic findings, repeated isolation of NTM organisms, the location from which NTM is cultured, and biopsy findings allow clinicians to distinguish between colonization and clinically significant infection.¹¹

M. gordonae is a Runyon group II organism considered to have limited pathogenicity and rare association with disease.^22,38,48 Detection of this organism in human patients most often is considered to represent transient colonization or specimen contamination.² The sole report of M. gordonae colonization in squirrel monkeys was associated with no gross or histologic lesions.⁴³ M. kansasii is a slow-growing, Runyon group I organism.³⁸ This species is thought to be more pathogenic relative to other NTM species, with approximately 50% to 88% of isolates associated with human clinical disease.^4,31 Consistent with this pattern, M. kansasii was cultured from tissue demonstrating granulomatous lymphadenitis, whereas we detected M. gordonae in the feces of healthy marmosets.

Granulomatous lymphadenitis was the most common histopathologic lesion that we observed in the current study. Ziehl– Neelsen and Fite–Faraco staining failed to reveal acid-fast organisms. Despite this result, M. kansasii could be cultured from these lesions. Acid-fast staining for detection of mycobacterial organisms has limited sensitivity, failing to detect organisms when fewer than 10⁴ bacilli are present per slide.⁴⁷ In addition, modification of the mycobacterial cell wall in persistent infection has been associated with reduced sensitivity of the Ziehl–Neelsen stain despite stable numbers of mycobacterial organisms.⁴⁰ Consistent with our findings, publications documenting M. kansasii-associated granulomatous disease in Old World and New World nonhuman primates reported an infrequent or a lack of acid-fast bacilli detection.^5,21

This case series documents an association between colonization with NTM species and the observation of suspect TST reactions in common marmosets. The significant increase in M. gordonae detection in the feces of TST suspect animals likely represents intestinal colonization, which may indicate exposure or susceptibility to NTM. Although further study is required, the presence of clinical disease as evidenced by histologic lesions is most likely associated with infection by the more pathogenic of the 2 organisms, M. kansasii. Although clinicians should always maintain a high index of suspicion for MTC, colonization with NTM organisms should be considered as a cause for equivocal or positive TST reactions in common marmosets.