There is significant relationship between Beijing genotype family strains and resistance to the first-line anti-tuberculosis drugs in the Iranian population

Dear Editor,

Tuberculosis is one of the most important global concerns, as it is considered as the second leading cause of death due to the infectious diseases after HIV. According to the WHO reports in 2020, it is estimated that around 10 million people have affected to TB and 1.5 million deaths occurred due to that in 2019 [1,2].

Iran is also one of the countries in the Middle-East region with a population of about 82 million; incidence of TB in Iran is estimated at about 11,000 people. Tuberculosis control and monitoring in Iran is very important due to its proximity to the countries such as Azerbaijan and Armenia (both high MDR-TB) and Afghanistan and Pakistan (with high-TB burden) [1], [2], [3]. One of the most important strategies in designing the regional programs for the control and prevention of tuberculosis is the identification and tracking of community circulating Mycobacterium tuberculosis (Mtb) strains by the molecular epidemiology techniques [3,4].

According to the literature review, the most common Mtb families in Iran are Beijing (Beijing / W), Haarlem (HaarlemI / HaarlemII and New-1), Delhi/Cas, East-Asian Indian, while TUR, Ural, LAM and bovis are less separable [5,6].

Evidence has been suggested in recent years about the relationship of drug resistance and some Mtb lineages [5], [6], [7]. Thus, this study was performed aiming to determine the frequency of dominant lineages in Iran to evaluate the possible relationship of each of the Beijing, Delhi/Cas, Haarlem and EAI lineages with the first-line drug resistance, especially isoniazid, rifampin and MDR-TB (Multi-drug resistant TB).

In order to fulfil the study, we attempted to investigate all the required studies of the Iranian population by searching PubMed, Scopus, Embase, Cochrane library, Google Scholar and Iranian databases Iranmedex, SID and ISC by March 2020. The search was done with the MeSH keywords including “Mycobacterium tuberculosis”, “TB”, “drug-resistance”, and “genotyping”. The inclusion criteria included: 1- Studies involving the prevalence of our genotypes in susceptible isolates and drug resistance, and 2- Studies in which genotyping was determined by the standard techniques of Spoligotyping, MIRU-VNTR, IS6110 RFLP, and Whole genome sequencing (WGS); and case reports, repetitive sample studies with, and nonclinical studies were included. Finally, the probable relation between the infection and each of the Beijing, Delhi/Cas, Haarlem and EAI genotypes with first-line drug resistance was measured by the Odds Ratio (ORs) with 95% confidence intervals and the Egger weighted regression method was also determined [8].

Based on our investigation, 28 original articles including Persian and English articles were included in the present quantitative analysis (Table 1). In the present study, 6671 Mtb strain data were evaluated. The studies were limited to 2006–2019 and were conducted in the areas including Tehran (National Research Institute of Tuberculosis and Lung Disease), Tabriz, Golestan, Kermanshah and Mashhad.

Table 1.

Characteristics of included studies.

First author	Publication year	City	Patients	MTB strains	MTB genotypes distribution								Typing method	Ref
					Beijing		CAS/Delhi		Haarlem		East African-Indian
					DR	Total	DR	Total	DR	Total	DR	Total
Ramazanzadeh	2006	NRITLD	345	195	195	195	0	0	0	0	0	0	Spoligotyping	9
Farnia	2006	NRITLD	3812	1074	12	12	11	11	44	44	10	10	Spoligotyping	10
Farnia	2007	NRITLD	NA	31	0	0	2	2	13	13	14	14	Spoligotyping	11
Amirmozafari	2007	NRITLD	NA	439	27	68	0	38	0	149	0	31	Spoligotyping	12
Farnia	2008	NRITLD	NA	258	1	12	0	49	8	20	0	58	Spoligotyping	13
Dousstdar	2008	NRITLD	NA	30	9	9	12	12	7	7	1	2	Spoligotyping	14
Doustdar	2008	NRITLD	NA	50	15	15	0	0	10	10	0	0	IS6110 RFLP	15
Doustdar	2009	NRITLD	NA	34	7	7	10	10	5	5	8	8	IS6110 RFLP	16
Rohani	2009	Mashhad	NA	113	5	8	0	0	0	0	0	0	Spoligotyping	17
Tajeddin	2009	NRITLD	NA	238	9	13	21	76	10	66	13	55	Spoligotyping	18
Velayati	2009	NRITLD	NA	146	5	5	4	4	9	9	5	5	MIRU-VNTR	19
Ahmadi	2009	NRITLD	238	238	9	13	21	66	10	66	13	55	Spoligotyping	20
Mozafari	2010	NRITLD	NA	105	7	20	NA	22	NA	2	0	0	Spoligotyping	21
Merza	2010	NRITLD	3812	1074	6	12	NA	181	NA	44	NA	227	Spoligotyping	22
Jafarian	2012	NRITLD	NA	96	1	3	7	7	15	36	0	0	MIRU-VNTR	23
Haeili	2013	NRITLD	NA	291	3	3	5	70	0	3	0	4	Spoligotyping	24
Sharifipour	2014	NRITLD	NA	190	5	11	8	37	8	72	0	0	Spoligotyping	25
Varahram	2014	NRITLD	151	151	10	14	4	23	12	31	14	42	Spoligotyping	26
Kardan Yamchi	2015	NRITLD	NA	31	8	8	2	2	1	1	1	1	Spoligotyping	27
Mohajeri	2016	Kermanshah	523	146	3	15	0	0	0	0	0	0	Multiplex PCR	28
Rezaei	2016	NRITLD	NA	20	3	3	3	3	10	10	0	0	Spoligotyping	29
Khanipour	2016	NRITLD	NA	723	9	9	1	1	11	11			Spoligotyping	30
Khosravi	2017	NRITLD	NA	88	9	9	0	0	3	3	0	0	IS6110 RFLP	31
Mansoori	2018	Golestan	11,807	164	5	19	3	31	0	2	0	0	MIRU-VNTR	32
Zaniani	2018	Isfahan	33	18	7	7	4	4	0	0	1	1	MIRU-VNTR	33
Vaziri	2019	NRITLD	13,892	606	14	14	7	7	0	0	0	0	WGS	34
Kazemian	2018	NRITLD	NA	34	11	11	4	4	0	0	1	1	Spoligotyping	SID
Afaghi-Gharamaleki	2019	Tabriz	125	115	3	7	0	1	0	0	0	0	MIRU-VNTR	35

NRITLD: National Research Institute of tuberculosis and lung disease; WGS: whole genome sequencing.

The studied Mtb strains included two groups of susceptible and drug-resistant TB, with drug-resistant TB strains being resistant to the antibiotics such as isoniazid, rifampin, streptomycin, ethambutol and pyrazinamide.

In general, the frequency of Beijing (22.5%), Delhi/Cas (66.7/66.7%), Haarlem (264/60.2) and EAI (22.3/514.5%) lineages were evaluated, while the distribution of lineages in first-line drug resistant strains were evaluated as Beijing (76.2%/398%), Delhi/Cas (19 / 129.5%), Haarlem (29/17.1%) and EAI (16/81%). The distribution of Mtb lineages between susceptible and drug-resistant TB is demonstrated in Fig. 1.

Fig. 1.

Distribution of Iranian Predominant Mycobacterium tuberculosis lineages between susceptible and drug-resistant Mtb strains.

Based on statistical analysis by our random effects model, a significant relationship was observed between the infection with Mtb Beijing strains and drug resistance (ORs: 1.54 (with 95% CIs); 0.7–3.5; Q = 35.94; I² = 61.04; P = 0.02). However, no significant relationship was observed between the infection with Delhi/Cas families (0.17; 0.11–0.22; Q = 131.4; I² = 84.0; P = 0.00), Harrlem (0.18; 0.1–0.2; Q = 126.5; I² = 84.9; EAI (0.15; 0.0–0.2; Q = 70.3; I² = 81.5; P = 0.00) and drug resistance. Moreover, there was a significant relationship between first line drugs with Mtb Beijing strains and resistance to isoniazid and rifampin (2.19; 1.0–4.2; Q = 77.6; I² = 88.4; P = 0.027 and 1.60; 0.75–3.41; Q = 59.1; I² = 89.84; P = 0.2 respectively). Thus, according to the present meta-analysis results, the Delhi/Cas, Haarlem, and Beijing families were the three most common lineages in the tuberculosis patients in Iran, respectively, while Beijing was evaluated with the most significant difference amongst DR-TB strains. We also found a significant relationship between the infection with Beijing Mtb strains and resistance to isoniazid and rifampin in the Iranian patients.

The Beijing tuberculosis family originated from Beijing, China; it was first introduced by Soolingen et al. (1995) as a predominant tuberculosis genotype of East Asia but has worldwide distribution and global distribution today [37,38]. The incidence of Beijing genotypes in recent years has been estimated at between 50% (East Asian countries) to 8.9% (Western countries) [6,38]. According to the review of the literature, the Beijing genotype is one of the predominant tuberculosis family in the Middle-East countries, especially in Iran [5,6,39]. Based on our analysis, the Beijing genotypes were the third most common genotype in Iran. Numerous studies have been published in recent years on the relationship between the Beijing TB genotypes and drug resistance [39], [40], [41]. Based on WGS (whole genome sequencing) data, the presence of multiple SNPs in the katG, rpoB, embB, rpsL, rrs, eis, gyrA and gyrB genes compared to other genotypes could be effective in increasing MIC relative to the anti-tuberculosis drugs [42]. There have been several studies, so far, on the relationship of the infection with Beijing genotypes and drug resistance. However, some other studies contradict this phenomenon [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45]. According to Ramazanzadeh et al. (2014) about 80% of drug resistant Mtb isolates from the Iranian patients belonged to the Beijing genotype. Also, in the study by Tarashi et al. (2017), amongst the MDR-TB isolates of Iranian patients the frequency of Beijing and Haarlem genotypes were 19.3% and 18.7%, respectively [5,6]. We also showed in this study that the Beijing genotype was the third most common genotype in the Iranian tuberculosis (TB) patients, and the infection with the Beijing family had significant relationship with the resistance to isoniazid and rifampin. Farnia et al. (2006) showed that Haarlem I and Beijing genotypes are the most common TB lineages of the Afghan population residing in Iran, who are affected to MDR-TB [10].

Other similar studies have also found that the Beijing genotype has been isolated from a significant population of TB patients in Iran neighbouring countries, especially Afghanistan and Pakistan. Hence, in order to prevent the spread of Mtb Beijing family infection, it seems that designing the guidelines has a significant impact on reducing the DR -TB cases in the Iranian population [5,6,39].

Ethical

Ethical Statement is not applicable for this manuscript.