TB-PRACTECAL Trial—Evidence for a shorter, safer, more effective treatment for drug-resistant tuberculosis
Drug-resistant tuberculosis (DR-TB) remains an especially deadly form of the ancient scourge of TB, while current treatments are long, toxic, and ineffective for half of all patients. Aiming to change this unacceptable status quo, in the mid-2010’s MSF and partners launched three clinical trials to test novel regimens containing the first new TB drugs in decades.
On 22 December 2022 the New England Journal of Medicine published findings from TB-PRACTECAL, a three-country randomized controlled trial, showing that a shorter regimen is safer and cured 89% of DR-TB patients, compared with 52% on the standard of care. These findings have already been incorporated into the World Health Organization’s new TB treatment guidelines. A separate study shows that the new regimen is also more cost-effective.
Alongside these results the content collection linked below highlights other aspects of the trial, from community engagement strategies that helped shape TB-PRACTECAL to setbacks arising from the Covid-19 pandemic. It also examines urgent challenges in scaling up access to these life-saving drugs, including affordability and patent barriers.
11 result(s)
Journal Article > ResearchFull Text
Lancet Respir Med. 1 February 2024; Volume 12 (Issue 2); 117-128.
Nyang'wa BT, Berry C, Kazounis E, Motta I, Parpieva N, et al.
Lancet Respir Med. 1 February 2024; Volume 12 (Issue 2); 117-128.
BACKGROUND
Around 500,000 people worldwide develop rifampicin-resistant tuberculosis each year. The proportion of successful treatment outcomes remains low and new treatments are needed. Following an interim analysis, we report the final safety and efficacy outcomes of the TB-PRACTECAL trial, evaluating the safety and efficacy of oral regimens for the treatment of rifampicin-resistant tuberculosis.
METHODS
This open-label, randomised, controlled, multi-arm, multicentre, non-inferiority trial was conducted at seven hospital and community sites in Uzbekistan, Belarus, and South Africa, and enrolled participants aged 15 years and older with pulmonary rifampicin-resistant tuberculosis. Participants were randomly assigned, in a 1:1:1:1 ratio using variable block randomisation and stratified by trial site, to receive 36-80 week standard care; 24-week oral bedaquiline, pretomanid, and linezolid (BPaL); BPaL plus clofazimine (BPaLC); or BPaL plus moxifloxacin (BPaLM) in stage one of the trial, and in a 1:1 ratio to receive standard care or BPaLM in stage two of the trial, the results of which are described here. Laboratory staff and trial sponsors were masked to group assignment and outcomes were assessed by unmasked investigators. The primary outcome was the percentage of participants with a composite unfavourable outcome (treatment failure, death, treatment discontinuation, disease recurrence, or loss to follow-up) at 72 weeks after randomisation in the modified intention-to-treat population (all participants with rifampicin-resistant disease who received at least one dose of study medication) and the per-protocol population (a subset of the modified intention-to-treat population excluding participants who did not complete a protocol-adherent course of treatment (other than because of treatment failure or death) and those who discontinued treatment early because they violated at least one of the inclusion or exclusion criteria). Safety was measured in the safety population. The non-inferiority margin was 12%. This trial is registered with ClinicalTrials.gov, NCT02589782, and is complete.
FINDINGS
Between Jan 16, 2017, and March 18, 2021, 680 patients were screened for eligibility, of whom 552 were enrolled and randomly assigned (152 to the standard care group, 151 to the BPaLM group, 126 to the BPaLC group, and 123 to the BPaL group). The standard care and BPaLM groups proceeded to stage two and are reported here, post-hoc analyses of the BPaLC and BPaL groups are also reported. 151 participants in the BPaLM group and 151 in the standard care group were included in the safety population, with 138 in the BPaLM group and 137 in the standard care group in the modified intention-to-treat population. In the modified intention-to-treat population, unfavourable outcomes were reported in 16 (12%) of 137 participants for whom outcome was assessable in the BPaLM group and 56 (41%) of 137 participants in the standard care group (risk difference -29·2 percentage points [96·6% CI -39·8 to -18·6]; non-inferiority and superiority p<0·0001). 34 (23%) of 151 participants receiving BPaLM had adverse events of grade 3 or higher or serious adverse events, compared with 72 (48%) of 151 participants receiving standard care (risk difference -25·2 percentage points [96·6% CI -36·4 to -13·9]). Five deaths were reported in the standard care group by week 72, of which one (COVID-19 pneumonia) was unrelated to treatment and four (acute pancreatitis, suicide, sudden death, and sudden cardiac death) were judged to be treatment-related.
INTERPRETATION
The 24-week, all-oral BPaLM regimen is safe and efficacious for the treatment of pulmonary rifampicin-resistant tuberculosis, and was added to the WHO guidance for treatment of this condition in 2022. These findings will be key to BPaLM becoming the preferred regimen for adolescents and adults with pulmonary rifampicin-resistant tuberculosis.
Around 500,000 people worldwide develop rifampicin-resistant tuberculosis each year. The proportion of successful treatment outcomes remains low and new treatments are needed. Following an interim analysis, we report the final safety and efficacy outcomes of the TB-PRACTECAL trial, evaluating the safety and efficacy of oral regimens for the treatment of rifampicin-resistant tuberculosis.
METHODS
This open-label, randomised, controlled, multi-arm, multicentre, non-inferiority trial was conducted at seven hospital and community sites in Uzbekistan, Belarus, and South Africa, and enrolled participants aged 15 years and older with pulmonary rifampicin-resistant tuberculosis. Participants were randomly assigned, in a 1:1:1:1 ratio using variable block randomisation and stratified by trial site, to receive 36-80 week standard care; 24-week oral bedaquiline, pretomanid, and linezolid (BPaL); BPaL plus clofazimine (BPaLC); or BPaL plus moxifloxacin (BPaLM) in stage one of the trial, and in a 1:1 ratio to receive standard care or BPaLM in stage two of the trial, the results of which are described here. Laboratory staff and trial sponsors were masked to group assignment and outcomes were assessed by unmasked investigators. The primary outcome was the percentage of participants with a composite unfavourable outcome (treatment failure, death, treatment discontinuation, disease recurrence, or loss to follow-up) at 72 weeks after randomisation in the modified intention-to-treat population (all participants with rifampicin-resistant disease who received at least one dose of study medication) and the per-protocol population (a subset of the modified intention-to-treat population excluding participants who did not complete a protocol-adherent course of treatment (other than because of treatment failure or death) and those who discontinued treatment early because they violated at least one of the inclusion or exclusion criteria). Safety was measured in the safety population. The non-inferiority margin was 12%. This trial is registered with ClinicalTrials.gov, NCT02589782, and is complete.
FINDINGS
Between Jan 16, 2017, and March 18, 2021, 680 patients were screened for eligibility, of whom 552 were enrolled and randomly assigned (152 to the standard care group, 151 to the BPaLM group, 126 to the BPaLC group, and 123 to the BPaL group). The standard care and BPaLM groups proceeded to stage two and are reported here, post-hoc analyses of the BPaLC and BPaL groups are also reported. 151 participants in the BPaLM group and 151 in the standard care group were included in the safety population, with 138 in the BPaLM group and 137 in the standard care group in the modified intention-to-treat population. In the modified intention-to-treat population, unfavourable outcomes were reported in 16 (12%) of 137 participants for whom outcome was assessable in the BPaLM group and 56 (41%) of 137 participants in the standard care group (risk difference -29·2 percentage points [96·6% CI -39·8 to -18·6]; non-inferiority and superiority p<0·0001). 34 (23%) of 151 participants receiving BPaLM had adverse events of grade 3 or higher or serious adverse events, compared with 72 (48%) of 151 participants receiving standard care (risk difference -25·2 percentage points [96·6% CI -36·4 to -13·9]). Five deaths were reported in the standard care group by week 72, of which one (COVID-19 pneumonia) was unrelated to treatment and four (acute pancreatitis, suicide, sudden death, and sudden cardiac death) were judged to be treatment-related.
INTERPRETATION
The 24-week, all-oral BPaLM regimen is safe and efficacious for the treatment of pulmonary rifampicin-resistant tuberculosis, and was added to the WHO guidance for treatment of this condition in 2022. These findings will be key to BPaLM becoming the preferred regimen for adolescents and adults with pulmonary rifampicin-resistant tuberculosis.
Journal Article > ResearchFull Text
N Engl J Med. 22 December 2022; Volume 387 (Issue 25); 2331-2343.
Nyang'wa BT, Berry C, Kazounis E, Motta I, Parpieva N, et al.
N Engl J Med. 22 December 2022; Volume 387 (Issue 25); 2331-2343.
BACKGROUND
In patients with rifampin-resistant tuberculosis, all-oral treatment regimens that are more effective, shorter, and have a more acceptable side-effect profile than current regimens are needed.
METHODS
We conducted an open-label, phase 2–3, multicenter, randomized, controlled, noninferiority trial to evaluate the efficacy and safety of three 24-week, all-oral regimens for the treatment of rifampin-resistant tuberculosis. Patients in Belarus, South Africa, and Uzbekistan who were 15 years of age or older and had rifampin-resistant pulmonary tuberculosis were enrolled. In stage 2 of the trial, a 24-week regimen of bedaquiline, pretomanid, linezolid, and moxifloxacin (BPaLM) was compared with a 9-to-20-month standard-care regimen. The primary outcome was an unfavorable status (a composite of death, treatment failure, treatment discontinuation, loss to follow-up, or recurrence of tuberculosis) at 72 weeks after randomization. The noninferiority margin was 12 percentage points.
RESULTS
Recruitment was terminated early. Of 301 patients in stage 2 of the trial, 145, 128, and 90 patients were evaluable in the intention-to-treat, modified intention-to-treat, and per-protocol populations, respectively. In the modified intention-to-treat analysis, 11% of the patients in the BPaLM group and 48% of those in the standard-care group had a primary-outcome event (risk difference, -37 percentage points; 96.6% confidence interval [CI], -53 to -22). In the per-protocol analysis, 4% of the patients in the BPaLM group and 12% of those in the standard-care group had a primary-outcome event (risk difference, -9 percentage points; 96.6% CI, -22 to 4). In the as-treated population, the incidence of adverse events of grade 3 or higher or serious adverse events was lower in the BPaLM group than in the standard-care group (19% vs. 59%).
CONCLUSIONS
In patients with rifampin-resistant pulmonary tuberculosis, a 24-week, all-oral regimen was noninferior to the accepted standard-care treatment, and it had a better safety profile. (Funded by Médecins sans Frontières; TB-PRACTECAL ClinicalTrials.gov number, NCT02589782. opens in new tab.)
In patients with rifampin-resistant tuberculosis, all-oral treatment regimens that are more effective, shorter, and have a more acceptable side-effect profile than current regimens are needed.
METHODS
We conducted an open-label, phase 2–3, multicenter, randomized, controlled, noninferiority trial to evaluate the efficacy and safety of three 24-week, all-oral regimens for the treatment of rifampin-resistant tuberculosis. Patients in Belarus, South Africa, and Uzbekistan who were 15 years of age or older and had rifampin-resistant pulmonary tuberculosis were enrolled. In stage 2 of the trial, a 24-week regimen of bedaquiline, pretomanid, linezolid, and moxifloxacin (BPaLM) was compared with a 9-to-20-month standard-care regimen. The primary outcome was an unfavorable status (a composite of death, treatment failure, treatment discontinuation, loss to follow-up, or recurrence of tuberculosis) at 72 weeks after randomization. The noninferiority margin was 12 percentage points.
RESULTS
Recruitment was terminated early. Of 301 patients in stage 2 of the trial, 145, 128, and 90 patients were evaluable in the intention-to-treat, modified intention-to-treat, and per-protocol populations, respectively. In the modified intention-to-treat analysis, 11% of the patients in the BPaLM group and 48% of those in the standard-care group had a primary-outcome event (risk difference, -37 percentage points; 96.6% confidence interval [CI], -53 to -22). In the per-protocol analysis, 4% of the patients in the BPaLM group and 12% of those in the standard-care group had a primary-outcome event (risk difference, -9 percentage points; 96.6% CI, -22 to 4). In the as-treated population, the incidence of adverse events of grade 3 or higher or serious adverse events was lower in the BPaLM group than in the standard-care group (19% vs. 59%).
CONCLUSIONS
In patients with rifampin-resistant pulmonary tuberculosis, a 24-week, all-oral regimen was noninferior to the accepted standard-care treatment, and it had a better safety profile. (Funded by Médecins sans Frontières; TB-PRACTECAL ClinicalTrials.gov number, NCT02589782. opens in new tab.)
Journal Article > ResearchFull Text
PLOS Glob Public Health. 7 December 2022; Volume 2 (Issue 12); e0001337.
Sweeney S, Berry C, Kazounis E, Motta I, Vassall A, et al.
PLOS Glob Public Health. 7 December 2022; Volume 2 (Issue 12); e0001337.
Current options for treating tuberculosis (TB) that is resistant to rifampicin (RR-TB) are few, and regimens are often long and poorly tolerated. Following recent evidence from the TB-PRACTECAL trial countries are considering programmatic uptake of 6-month, all-oral treatment regimens. We used a Markov model to estimate the incremental cost-effectiveness of three regimens containing bedaquiline, pretomanid and linezolid (BPaL) with and without moxifloxacin (BPaLM) or clofazimine (BPaLC) compared with the current mix of long and short standard of care (SOC) regimens to treat RR-TB from the provider perspective in India, Georgia, Philippines, and South Africa. We estimated total costs (2019 USD) and disability-adjusted life years (DALYs) over a 20-year time horizon. Costs and DALYs were discounted at 3% in the base case. Parameter uncertainty was tested with univariate and probabilistic sensitivity analysis. We found that all three regimens would improve health outcomes and reduce costs compared with the current programmatic mix of long and short SOC regimens in all four countries. BPaL was the most cost-saving regimen in all countries, saving $112-$1,173 per person. BPaLM was the preferred regimen at a willingness to pay per DALY of 0.5 GDP per capita in all settings. Our findings indicate BPaL-based regimens are likely to be cost-saving and more effective than the current standard of care in a range of settings. Countries should consider programmatic uptake of BPaL-based regimens.
Technical Report > Policy Brief
MSF Access Campaign
8 November 2022
TB was the leading cause of death from a single infectious agent until the COVID pandemic. The number of people newly diagnosed with TB in 2020 fell by 18% from the previous year due to disruptions to health systems and services caused by the pandemic, with only a partial recovery in 2021. As a result, in 2021, only one in three people with drug-resistant TB (DR-TB) received treatment for the disease.
However, since the onset of the pandemic, more effective and patient-friendly treatments and regimens for adults and children have become available to the TB community. Now more than ever there is a need to accelerate treatment and save more lives.
This Issue Brief – the eighth in this series – by Médecins Sans Frontières (MSF)’s Access Campaign, examines the current landscape and trends of DR-TB drug pricing and patents, and highlights challenges and opportunities to accelerate people’s access to lifesaving regimens that are shorter, all-oral and make use of the most effective medicines.
However, since the onset of the pandemic, more effective and patient-friendly treatments and regimens for adults and children have become available to the TB community. Now more than ever there is a need to accelerate treatment and save more lives.
This Issue Brief – the eighth in this series – by Médecins Sans Frontières (MSF)’s Access Campaign, examines the current landscape and trends of DR-TB drug pricing and patents, and highlights challenges and opportunities to accelerate people’s access to lifesaving regimens that are shorter, all-oral and make use of the most effective medicines.
Journal Blog > Perspective
BMJ Opinion (blog). 29 June 2022; Volume 377; o1598.
Jain L
BMJ Opinion (blog). 29 June 2022; Volume 377; o1598.
Journal Article > ResearchFull Text
Trials. 13 June 2022; Volume 23 (Issue 1); 484.
Berry C, du Cros PAK, Fielding K, Gajewski S, Kazounis E, et al.
Trials. 13 June 2022; Volume 23 (Issue 1); 484.
BACKGROUND
Globally rifampicin-resistant tuberculosis disease affects around 460,000 people each year. Currently recommended regimens are 9-24 months duration, have poor efficacy and carry significant toxicity. A shorter, less toxic and more efficacious regimen would improve outcomes for people with rifampicin-resistant tuberculosis.
METHODS
TB-PRACTECAL is an open-label, randomised, controlled, phase II/III non-inferiority trial evaluating the safety and efficacy of 24-week regimens containing bedaquiline and pretomanid to treat rifampicin-resistant tuberculosis. Conducted in Uzbekistan, South Africa and Belarus, patients aged 15 and above with rifampicin-resistant pulmonary tuberculosis and requiring a new course of therapy were eligible for inclusion irrespective of HIV status. In the first stage, equivalent to a phase IIB trial, patients were randomly assigned one of four regimens, stratified by site. Investigational regimens include oral bedaquiline, pretomanid and linezolid. Additionally, two of the regimens also included moxifloxacin (arm 1) and clofazimine (arm 2) respectively. Treatment was administered under direct observation for 24 weeks in investigational arms and 36 to 96 weeks in the standard of care arm. The second stage of the study was equivalent to a phase III trial, investigating the safety and efficacy of the most promising regimen/s. The primary outcome was the percentage of unfavourable outcomes at 72 weeks post-randomisation. This was a composite of early treatment discontinuation, treatment failure, recurrence, lost-to-follow-up and death. The study is being conducted in accordance with ICH-GCP and full ethical approval was obtained from Médecins sans Frontières ethical review board, London School of Hygiene and Tropical Medicine ethical review board as well as ERBs and regulatory authorities at each site.
DISCUSSION
TB-PRACTECAL is an ambitious trial using adaptive design to accelerate regimen assessment and bring novel treatments that are effective and safe to patients quicker. The trial took a patient-centred approach, adapting to best practice guidelines throughout recruitment. The implementation faced significant challenges from the COVID-19 pandemic. The trial was terminated early for efficacy on the advice of the DSMB and will report on data collected up to the end of recruitment and, additionally, the planned final analysis at 72 weeks after the end of recruitment.
Globally rifampicin-resistant tuberculosis disease affects around 460,000 people each year. Currently recommended regimens are 9-24 months duration, have poor efficacy and carry significant toxicity. A shorter, less toxic and more efficacious regimen would improve outcomes for people with rifampicin-resistant tuberculosis.
METHODS
TB-PRACTECAL is an open-label, randomised, controlled, phase II/III non-inferiority trial evaluating the safety and efficacy of 24-week regimens containing bedaquiline and pretomanid to treat rifampicin-resistant tuberculosis. Conducted in Uzbekistan, South Africa and Belarus, patients aged 15 and above with rifampicin-resistant pulmonary tuberculosis and requiring a new course of therapy were eligible for inclusion irrespective of HIV status. In the first stage, equivalent to a phase IIB trial, patients were randomly assigned one of four regimens, stratified by site. Investigational regimens include oral bedaquiline, pretomanid and linezolid. Additionally, two of the regimens also included moxifloxacin (arm 1) and clofazimine (arm 2) respectively. Treatment was administered under direct observation for 24 weeks in investigational arms and 36 to 96 weeks in the standard of care arm. The second stage of the study was equivalent to a phase III trial, investigating the safety and efficacy of the most promising regimen/s. The primary outcome was the percentage of unfavourable outcomes at 72 weeks post-randomisation. This was a composite of early treatment discontinuation, treatment failure, recurrence, lost-to-follow-up and death. The study is being conducted in accordance with ICH-GCP and full ethical approval was obtained from Médecins sans Frontières ethical review board, London School of Hygiene and Tropical Medicine ethical review board as well as ERBs and regulatory authorities at each site.
DISCUSSION
TB-PRACTECAL is an ambitious trial using adaptive design to accelerate regimen assessment and bring novel treatments that are effective and safe to patients quicker. The trial took a patient-centred approach, adapting to best practice guidelines throughout recruitment. The implementation faced significant challenges from the COVID-19 pandemic. The trial was terminated early for efficacy on the advice of the DSMB and will report on data collected up to the end of recruitment and, additionally, the planned final analysis at 72 weeks after the end of recruitment.
Conference Material > Video
Berry C, Motta I, Kazounis E, Fielding K, Dodd M, et al.
MSF Scientific Days International 2022. 7 June 2022
Journal Blog > Perspective
PLoS Blogs. 10 May 2022
Berry C
PLoS Blogs. 10 May 2022
Journal Article > ResearchFull Text
Trials. 4 December 2021; Volume 22; 881 .
Wharton-Smith A, Horter SCB, Douch E, Gray NSB, James N, et al.
Trials. 4 December 2021; Volume 22; 881 .
BACKGROUND
Addressing the global burden of multidrug-resistant tuberculosis (MDR-TB) requires identification of shorter, less toxic treatment regimens. Médecins Sans Frontières (MSF) is currently conducting a phase II/III randomised controlled clinical trial, to find more effective, shorter and tolerable treatments for people with MDR-TB. Recruitment to the trial in Uzbekistan has been slower than expected; we aimed to study patient and health worker experiences of the trial, examining potential factors perceived to impede and facilitate trial recruitment, as well as general perceptions of clinical research in this context.
METHODS
We conducted a qualitative study using maximum variation, purposive sampling of participants. We carried out in-depth interviews (IDIs) and focus group discussions (FGDs) guided by semi-structured topic guides. In December 2019 and January 2020, 26 interviews were conducted with patients, Ministry of Health (MoH) and MSF staff and trial health workers, to explore challenges and barriers to patient recruitment as well as perceptions of the trial and research in general. Preliminary findings from the interviews informed three subsequent focus group discussions held with patients, nurses and counsellors. Focus groups adopted a person-centred design, brainstorming potential solutions to problems and barriers. Interviews and FGDs were audio recorded, translated and transcribed verbatim. Thematic analysis, drawing on constant comparison, was used to analyse the data.
RESULTS
Health system contexts may compete with new approaches especially when legislative health regulations or policy around treatment is ingrained in staff beliefs, perceptions and practice, which can undermine clinical trial recruitment. Trust plays a significant role in how patients engage with the trial. Decision-making processes are dynamic and associated with relationship to diagnosis, assimilation of information, previous knowledge or experience and influence of peers and close relations.
CONCLUSIONS
This qualitative analysis highlights ways in which insights developed together with patients and healthcare workers might inform approaches towards improved recruitment into trials, with the overall objective of delivering evidence for better treatments.
Addressing the global burden of multidrug-resistant tuberculosis (MDR-TB) requires identification of shorter, less toxic treatment regimens. Médecins Sans Frontières (MSF) is currently conducting a phase II/III randomised controlled clinical trial, to find more effective, shorter and tolerable treatments for people with MDR-TB. Recruitment to the trial in Uzbekistan has been slower than expected; we aimed to study patient and health worker experiences of the trial, examining potential factors perceived to impede and facilitate trial recruitment, as well as general perceptions of clinical research in this context.
METHODS
We conducted a qualitative study using maximum variation, purposive sampling of participants. We carried out in-depth interviews (IDIs) and focus group discussions (FGDs) guided by semi-structured topic guides. In December 2019 and January 2020, 26 interviews were conducted with patients, Ministry of Health (MoH) and MSF staff and trial health workers, to explore challenges and barriers to patient recruitment as well as perceptions of the trial and research in general. Preliminary findings from the interviews informed three subsequent focus group discussions held with patients, nurses and counsellors. Focus groups adopted a person-centred design, brainstorming potential solutions to problems and barriers. Interviews and FGDs were audio recorded, translated and transcribed verbatim. Thematic analysis, drawing on constant comparison, was used to analyse the data.
RESULTS
Health system contexts may compete with new approaches especially when legislative health regulations or policy around treatment is ingrained in staff beliefs, perceptions and practice, which can undermine clinical trial recruitment. Trust plays a significant role in how patients engage with the trial. Decision-making processes are dynamic and associated with relationship to diagnosis, assimilation of information, previous knowledge or experience and influence of peers and close relations.
CONCLUSIONS
This qualitative analysis highlights ways in which insights developed together with patients and healthcare workers might inform approaches towards improved recruitment into trials, with the overall objective of delivering evidence for better treatments.
Journal Article > LetterFull Text
Trials. 29 May 2021; Volume 22 (Issue 1); 371.
Nyang'wa BT, LaHood AN, Mitnick CD, Guglielmetti L
Trials. 29 May 2021; Volume 22 (Issue 1); 371.
When 2020 opened, approximately 11 million new tuberculosis (TB) cases and nearly 1.5 million TB-related deaths were predicted during the year. And, the gap between required and available research and development resources for TB was estimated at more than 1 billion USD. COVID-19, the global pandemic of disease caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has surely widened that gap by worsening the TB pandemic. Worryingly, the economic devastation wrought by COVID-19 portends further dramatic reductions in funds for other health research, including for multidrug-resistant tuberculosis (MDR-TB). Such constraints will prevent the use of extant TB research infrastructure to support cross-disease benefits in the struggle against COVID-19 and future pandemics.
Journal Blog > Perspective
BMJ Opinion (blog). 9 November 2018
Douch E
BMJ Opinion (blog). 9 November 2018
