Background

Multidrug-resistant tuberculosis (MDR-TB) is a major global health problem with treatment success rate less than 60% [1]. The 2019 WHO consolidated guidelines recommended three kinds of therapeutic options (all-oral long/short or injection included) for countries and programs treating MDR-TB [1]. However, varied status in different areas or countries resulted in varied outcomes, depending on different factors such as financial support, management, protocol, drug quality, newly included drugs, patient compliance and tolerance. Therefore, how to achieve the optimal treatment outcome for MDR-TB and ensure its high safety has become the most important issue for clinicians. The all-oral regimen containing bedaquiline (BDQ) instead of injectable agents recommended in 2019 guideline [2] may be effective but difficult to be implemented in resource-poor areas due to its high price. It was pointed out that the explicit recommendations for the use of Linazolamide (LZD) and BDQ might bring some disadvantages to the global TB control program [3]. The ideal treatment for MDR-TB should be effective, with fewer adverse events (AEs) and affordable, especially in resource-poor areas with high TB burden [4].

According to the WHO 2019 report, the treatment coverage rate of MDR-TB in China was only 13.6% [1]. Delayed drug sensitivity test (DST), unaffordable treatment costs due to long courses or expensive drugs and AEs were the main reasons [11] at the end of the 12th month: among which 26 (72%, 26/36) patients were newly treated, and 25 (32%, 25/78) were retreated, p value was 0.00.

All patients who completed the course of treatment were followed for 1 year and no recurrence was found.

Safety and adverse events (AEs)

The investigators recorded 42 AEs occurrences in 30 of 114 patients (30/114, 26.3%). Clinicians rated most AEs as mild or moderate (n = 40 AEs, 40/42) and no one was caused by Pa (0/114, 0%). The incidence of AEs were significantly lower in newly treated patients (6/36, 16.67%) than those in retreated patients (24/78, 30.77%).

Among them, 22 AEs occurrences in 15 patients (13.16%, 15/114) experienced a change in treatment regimen. Among them, 12 AEs occurrence in 10 patients associated with Ak due to mild hearing loss (n = 5) or mild renal dysfunction (n = 7). 6 AEs occurrences in 3 patients associated with Z due to transaminase elevation (n = 3, alanine aminotransferase were elevated to 2 times higher than the upper limit of the normal range) and gout (n = 3). 4 AEs occurrences in 2 patients associated with Pto due to transaminase elevation (n = 2, alanine aminotransferase were elevated two times higher than the upper limit of the normal) and gastrointestinal reaction (n = 2). All AEs above were improved after discontinuation of the suspected drugs. 9 (60%, 9/15) of these patients returned to the discontinued drug after dosage adjustment and completed the course.

4 AEs in 2 patients (1.75%, 2/114) were reported to be severe enough to require permanent discontinuation of the suspected drug: 1 case (0.88%, 1/114) was associated with Ak by hearing loss (n = 1) and mild renal insufficiency (n = 1). Allergic rash (n = 1) and gastrointestinal reaction (n = 1) caused by Mfx in 1 case (0.88%, 1/114). Patient with allergic skin rash were unable to tolerate and abandoned the treatment.

As the rest 16 AEs occurrences in 13 patients (11.40%, 13/114) were improved after symptomatic treatment, and no treatment plan was changed: 4 AEs occurrences in 4 patients suspended Ak due to mild dizziness (n = 4); 12 AEs occurrences in 9 patients suspended Z and Pto due to mild transaminase elevation (n = 8, alanine aminotransferase elevated 2 times lower than the upper limit of the normal range.) and gastrointestinal reaction (n = 4), See in hot map of AE in Fig. 3.

Fig. 3
figure 3

Heatmap of the AEs composition (Lines) of the 30 patients (Columns). 42 AEs were found in 30 patients, and some patients may have more than one AE. The X axis represents 30 patients, the Y axis represents the type of the AEs, and the different colors represent the degree of the AEs. There will be several clusters for each patient

Full size image

Among the 30 patients with AE, treatment success was in 25 patients (83.3%) including 5 patients from newly treated group and 20 patients from retreated group; unfavorable outcome was in 5 patients (17.7%), including 2 cases withdrawing treatment due to AE of the drugs, 1 cases withdrawing the treatment without specific reason, 1 case getting failure and 1 case loss of follow-up.

Discussion

In the present study, we made up a regimen for the treatment on MDR-TB mainly according to guidelines of WHO 2016 [16]. We included MDR-TB patients with no resistance to FQs or injectable agents based on MIC DST. The importance of DST for individualized MDR-TB treatment has been well established, with resistance to key drugs associated with poor treatment outcomes [17]. The correlation between the MIC and the treatment outcomes of MDR-TB has also been verified by several studies [18, 19]. In the present study, the results showed that the overall favorable treatment rate was 79.8%, especially as high as (91.7%, 33/36) in newly treated patients. The incidence of AEs were also low, especially in newly treated patients (6/36, 16.67%) and the majority of AEs (n = 40 AEs, 40/42) were mild-to-moderate. The results demonstrated the high efficacy and safety of this regimen against specific MDR-TB patients. At the same time, the clinical guidance value of the DST of anti-tuberculosis drugs screened by MIC had positive guiding value for clinical practice.

Previous study shown that for patients without HIV infection, sputum culture negative conversion time can be regarded as an essential signal for successful treatment of MDR TB patients [20]. In the present study, we recorded sputum culture results every 3 months and plotted a Kaplan–Meier curve for patients with sputum culture cumulative conversion over the course of 1 year of treatment. The results showed that the sputum negative conversion rate of newly treated patients was significantly higher than that of retreated patients. Therefore, this regimen may be more suitable for newly treated patients. In addition, clinicians should pay close attention to the retreated patients who may have a worse prognosis.

The Bangladesh short range regimen recommended by the 2016 guidelines is suggested for treating MDR patients without FQs and second-line injectable agents resistance, which consisted of an intensive period of 4 to 6 months with seven drugs (Ak–FQs–Pto–Clofazimine (Cfz)–Z-high dose INH-E(Ethambutol) followed by a five-month course of FQs–Cfz–Z–E [21]. But Cfz is expensive in China (about $ 400/month) and may be unacceptable to most Asian patients due to the skin pigmentation, especially young women. In the present study, under inclusion conditions similar to the short range regimen: we replaced the high dose INH with Pa and replaced Cfz with Cs. The results indicated high favorable treatment outcome rate and the entire treatment regimen costs at only around $300 per month, while the short range regimen costs $450 per month and the all-oral regimen recommended by the 2019 WHO guideline costs more than $2000 per month in China. For resource-limited areas, the cost of treatment is an important determinant of patient compliance. Costly drugs, even if effective, may cause patients to discontinue treatment, which not only lead to the spread of drug-resistant TB, but can also lead to more complex drug resistance in individuals. China is a high TB burden country with high financial burden of health care. Therefore, BDQ had not been widely used at present for its price. On the other hand, the WHO’s clear recommendations on the use of LZD and BDQ could prove a double-edged sword for global TB control programmes [3]. Intuitively, the advantages of including LZD and BDQ in standard protocols for all types of MDR-TB may be more conducive to programme implementation and less likely to require DST. The main drawback, however, may be concerns about patient’s safety and tolerability. Lzd has significant long-term side effects as an ultra-broad-spectrum antibiotic and about 30–40% of patients stop linezolid treatment because of AEs [22]. Therefore, in the present study, patients were selected with inclusion criteria, and only FQs was selected as the included drug among class A drugs.

The rate of INH acetylated is controlled by genetics. Once acetylated, INH is ineffective as an antibiotic against TB bacilli. Rapid acetylation of INH may lead to low serum concentrations of anti-TB drugs, increasing the risk of treatment failure. Most Asians are of the fast-metabolizing type [23]. It had been reported that in the INH-resistant organisms studied, about 50% of INH MIC belong to the category of low concentration resistance with MIC at 0.1–1.0 ug/mL [24, 25]. A study from China showed that among 109 INH-resistant isolates, only 11.9% and 19.3% showed resistance to PAS and Pa, respectively [26]. Pa is a chemical synthesis of isoniazid (INH) and paminosalicylic acid (PAS). PAS effectively delays and blocks the acetylation of INH in vivo. Pa maintains high, prolonged concentration of INH in the blood and reduces toxicity to the liver. It not only enhances the bactericidal action of the drug, but also delays the generation of bacterial resistance. In the present study, patients infected with low concentration INH resistant strains were included, and MIC values showed that all strains were sensitive to Pa. Another advantage of Pa is its low price ($22 per month). Its safety was reflected in the incidence of AEs (0%).

Cycloserine (Cs) in group B has good antibacterial activity and the price in China is much lower than Cfz ($250 vs $400 per month). Due to its low drug resistance rate and low cross-resistance with other anti-TB drugs, it is often used as a good alternative drug for MDR-TB [27, 28]. The present study also verified its safety and efficacy.

Second-line injections (SLIs) were once one of the core drug groups in treatment of MDR-TB [16]. However, existing studies have shown that SLIs has high AEs and often leads to withdrawal of the treatment [29]. A retrospective study of 25 countries [30], according to the results of injection therapy (Ak) were better than no injections, but in analysis of comprehensive treatment results, patients who received the injection did worse than those who did not receive the injection. The results provided evidence for 2019 guidelines on the recommendation regarding the use of injectable agents [1]. But the article also explained that Ak may be the most widely used injectable drug due to its price and tended to be used in patients with the worst resistance patterns, which may be one of the reasons for poor outcomes rather than the problem with the drug itself. However, the present study showed that 18 AEs occurrences in 15 (13.16%, 15/114) patients suspicious of Ak due to mild hearing loss or mild renal dysfunction. But only 1 (6.67%, 1/15) patient had Ak permanently disabled. The results suggest that the Chinese population could be moderately tolerant to SLIs and that the cheap drug is certainly effective for certain populations.

Our study still had some limitations, including the relatively small sample size and the short follow-up period. Existing study shown that 3% of patients experienced MDR-TB recurrence during an average follow-up period of 4.8 years [31]. Therefore, it may be necessary to enroll more patients and follow up longer time to validate our conclusions.

Conclusions

The regimen in the present study had the following characteristics: highly effective with favorable treatment rate in newly treatment patients reached 91%; Pa instead of high-dose INH and CS instead of Cfz in the treatment of specific MDR-TB populations may be more suitable for Chinese; It proved that the AEs of SLIs are controllable in Chinese population and inexpensive; 72% (26/36) newly treated patients achieved the standard of withdraw medicine at the end of the 12th month. These characteristics suggested that the regimen could be widely used in China, even other resource-poor parts of Asia. Further research on the possibility of short-course treatment in resource-poor areas with high TB burden could be expected.