Correspondence regarding “Delamanid for rifampicin-resistant tuberculosis: a retrospective study from South Africa”
- Erika Mohr-Holland1,2⇑,
- Anja Reuter1,
- Jennifer Hughes1,
- Johnny Daniels1,
- Busisiwe Beko1,
- Goodman Makhanda3,
- Virginia De Avezedo4,
- Yulene Kock5,
- Helen Cox6,
- Jennifer Furin7,
- Laura Trivino Duran1,
- Petros Isaakidis2 and
- Gabriella Ferlazzo2
- 1Médecins Sans Frontières, Operational Centre Brussels (OCB), Khayelitsha Project and Coordination Office, Cape Town, South Africa
- 2Médecins Sans Frontières, South African Medical Unit (SAMU), Cape Town, South Africa
- 3South African Medical Research Council, Cape Town, South Africa
- 4City of Cape Town Health Department, Cape Town, South Africa
- 5National Department of Health, Pretoria, South Africa
- 6Institute of Infectious Disease and Molecular Medicine and Wellcome Centre for Infectious Diseases Research in Africa, University of Cape Town, Cape Town, South Africa
- 7Department of Global Health and Social Medicine, Harvard Medical School, Boston, MA, USA
- Erika Mohr-Holland, Médecins Sans Frontières (MSF) and MSF Southern African Medical Unit, The Isisivana Center, 1 Julius Tsolo Road, Khayelitsha, 7784, South Africa. E-mail: msfocb-khayelitsha-drtb-epi{at}brussels.msf.org
Abstract
Delamanid (Dlm) containing regimens to treat RR-TB are effective, but more data on Dlm use is needed. Dlm remains an important treatment option in programmatic settings with high rates of HIV co-infection and in persons with limited treatment options. https://bit.ly/3cVLeUq
To the Editor:
Here we report on the final outcomes for the cohort of 103 patients, 77% HIV-positive with a median (interquartile range, IQR) CD4 count of 141 (61–252) cells per mm3, from Khayelitsha, South Africa who initiated regimens containing delamanid (Dlm) for the treatment of rifampicin-resistant tuberculosis (RR-TB) from November 2015 to August 2017 [1]. The median (IQR) duration on Dlm was 6.3 (4.3–12.0) months; 54 (52%) patients received Dlm for more than 6 months and 32 (31%) also received a regimen containing bedaquiline (Bdq), 26 (81%) of whom harboured fluoroquinolone (Fq) resistant strains.
Overall, 57 (55%), 22 (21%), 14 (14%), six (6%), and four (4%) patients had treatment success (cured/completed treatment), loss to follow-up, death, failure of treatment, and transfer-out, respectively. The median (IQR) times to poor outcomes were as follows: loss to follow-up 10.4 (4.6–16.8) months, death 4.6 (3.3–14.3) months, and the failure of treatment 14.2 (9.2–16.7) months. Three (21%) and 10 (71%) of the deaths occurred within 3 and 6 months of treatment initiation (p=0.021). Four of the six cases with failure of treatment (all of whom were HIV-positive) had received Dlm due either to previous RR-TB treatment failure or limited therapeutic options; additionally, four cases received Dlm for >6 months due to delayed treatment response. Two of these patients had culture converted to negative and thereafter reconverted to positive while the remaining four never culture converted to negative. At the time in which the outcome of failure was assigned, four of the six cases were no longer on Dlm. There were no significant differences in final outcomes among patients with Fq-resistant compared to Fq-susceptible strains (66% versus 51%; p=0.33).
As we were unable to confirm what happened to the four patients transferred out, they were excluded from further analyses, leaving 78 and 21 HIV-positive and -negative patients. The rate of unsuccessful outcomes (loss to follow-up, death and failure of treatment) was higher among HIV-positive patients when compared to HIV-negative patients (n=38, 49% versus n=4, 19%; p=0.024). In the unadjusted analysis, male sex and HIV positivity, regardless of CD4 count, aside from those with an unknown CD4 count, were significantly associated with unsuccessful outcomes. After adjustment for potential confounders, male sex and HIV-positivity with a CD4 count <200 copies per mm3 remained significantly associated with unsuccessful outcomes (adjusted hazard ratio (aHR) 3.0, 95% CI 1.4–6.3; p=0.003; and aHR 3.8, 95% CI 1.2–11.6; p=0.020; respectively) (table 1). No factors were identified as significantly associated with successful outcomes in the adjusted analysis.
In total, there were 74 serious adverse events (SAEs) experienced by 31 (30%) patients; the overall incidence was 4.63 SAEs per 100 person months. Of the 74 SAEs, 25 (34%) were assessed as related to Dlm. The most common SAEs were QTcF prolongation (n=8 instances), anaemia (n=7 instances), meningitis (n=6 instances), seizures (n=6 instances), vomiting (n=5 instances), diarrhoea (n=3 instances), psychotic disorder (n=3 instances), and renal impairment (n=3 instances). The eight instances of QTcF prolongation reported as serious occurred in eight (8%) patients, of whom five (63%) and one (12%) were receiving the combination of Dlm, Bdq and clofazimine, and Dlm and Bdq, respectively. The frequency of SAEs reported among HIV-positive and -negative patients did not significantly differ (34% versus 17%; p=0.13). Dlm was withdrawn as a result of 13 SAEs in seven (7%) patients; 41 SAEs were recovered/resolved while 18 were fatal (two fatal SAEs were assessed as related to Dlm).
Delamanid was classified as a Category C drug in the recent World Health Organization recommendations and constitutes an important drug for those with limited therapeutic options receiving treatment for RR-TB [2]. While the rate of treatment success reported in this cohort is similar to the ones reported globally in RR-TB patients [3], it is relatively low when compared to success rates observed in other cohorts of patients treated with Dlm, mostly in low HIV-prevalence settings [4, 5]. Our data suggest that HIV infection still plays a leading role in driving unsuccessful outcomes among those treated for RR-TB, despite access to new drugs [3]. The high frequency of patients with advanced HIV disease may explain the high proportion of early deaths at 6 months (71%). Additionally, the higher frequency of loss to follow-up among HIV-positive patients highlights that the challenges associated with the dual burden of disease, for example pill burden and the occurrence of frequent non-serious side-effects, that might trump treatment completion [6]. Although second-line resistance is known to be associated with poorer outcomes, we did not observe differences in outcomes among patients with more complex resistance profiles, notably Fq resistance; this is possibly due to the fact that the vast majority (81%) received regimens containing both Bdq and Dlm. In line with previously reported results [7–9], our data suggest the effective role of the combination in improving outcomes among the most difficult to treat patients.
The RR-TB treatment outcomes among patients receiving Dlm in this cohort were in line with those reported nationally and globally; more programmatic data are needed to better understand the role of Dlm in the treatment of RR-TB [10]. Meanwhile, Dlm remains an important treatment option in programmatic settings with high rates of HIV co-infection and in persons with limited treatment options. Despite improved access to this novel agent, the rates of death and loss to follow-up in this cohort were high, particularly among those HIV co-infected, reflecting the challenges associated with the management of comorbidities and the prolonged duration of treatment for RR-TB. Broader and earlier use of new drugs, early diagnosis of advanced HIV disease and early initiation of appropriate antiretroviral therapy, as well as innovative psychosocial interventions, are needed to increase the chance to cure patients with RR-TB and decrease the number of preventable deaths.
Shareable PDF
Supplementary Material
This one-page PDF can be shared freely online.
Shareable PDF ERJ-00837-2020.Shareable
Footnotes
Conflict of interest: E. Mohr-Holland has nothing to disclose.
Conflict of interest: A. Reuter has nothing to disclose.
Conflict of interest: J. Hughes has nothing to disclose.
Conflict of interest: J. Daniels has nothing to disclose.
Conflict of interest: B. Beko has nothing to disclose.
Conflict of interest: G. Makanda has nothing to disclose.
Conflict of interest: V. De Azevedo has nothing to disclose.
Conflict of interest: Y. Kock has nothing to disclose.
Conflict of interest: H. Cox reports grants from Wellcome Trust, UK MRC and EDCTP, outside the submitted work.
Conflict of interest: J. Furin has nothing to disclose.
Conflict of interest: L. Trivino Duran has nothing to disclose.
Conflict of interest: P. Isaakidis has nothing to disclose.
Conflict of interest: G. Ferlazzo has nothing to disclose.
- Received March 24, 2020.
- Accepted June 9, 2020.
- Copyright ©ERS 2020