There are two live vaccine candidates in the global pipeline22 intended to replace BCG while being safer (Tables 1), namely VPM1002 and MTBVAC. VPM1002 is a recombinant Mycobacterium bovis rBCGΔureC::Hly+ whose safety and immunogenicity have been demonstrated in three clinical trials (CT). This candidate is currently in phase II/III in the clinical development pathway to check its efficacy in preventing the recurrence of TB (https://clinicaltrials.govNCT03152903).11,23 MTBVAC was generated from an Mtb strain and carries two independent attenuating mutations in the transcription factor phoP and the lipid biosynthesis gene fadD26, respectively. After showing its efficacy in experimental animal models and two phase I clinical trials in adults and children,24,25 it is now being evaluated in a phase Ib/IIa study (NCT02933281).

RUTI, DAR-901, M. vaccae and M. indicus pranii (MIP) are the four inactivated whole-cell vaccine candidates currently in the pipeline. The RUTI vaccine26 was primarily designed as a therapeutic vaccine,27 although it was subsequently found to be effective as a prophylactic in animal models.28 After a successful phase IIa trial in South Africa,29 it is now being tested as co-adjuvant treatment in MDR-TB patients (Phase IIa, NCT02711735). DAR-901 is a whole-cell vaccine manufactured in compliance with Good Manufacturing Practice from SRL172, which is derived from inactivated M. obuense (initially thought to be M. vaccae), and which was used in a successful phase III trial to show the efficacy and safety for the prevention of disseminated TB in HIV-positive patients (DARDAR study, NCT00052195).30 DAR-901 is now being tested in a phase II clinical trial as a booster to BCG for preventing infection with TB (NCT02712424). Another vaccine made from inactivated M. vaccae31 is being tested in a Prevention of Disease (PoD) phase III clinical trial in China in 10,000 Mtb-infected adults (NCT01979900) and as an adjunct to standard chemotherapy in the form of an oral pill (V7) in a phase III clinical trial in the Ukraine involving MDR-TB patients (NCT01977768). The MIP vaccine, which is used against leprosy and has already been tested as adjunct therapy in three TB clinical trials, will be soon evaluated together with VMP1002 in a phase III PoD trial in 19,000 household contacts of people with TB in India.32

Vaccines are commonly developed by generating constructs containing Mtb antigens (alone or in combination), which are delivered together with adjuvants or viral-vectored. The advantage of these candidates is that they are safer but they potentially have the drawback of being less immunogenic, as whole cell vaccines, due to their complexity, can induce conventional and non-conventional T cell and antibody responses and can train the immune system against TB as well as against other microorganisms, as has been shown for BCG.33 The primary objective of most vaccine candidates is either to increase the generation of protective lymphocyte cell subsets, to improve their presentation or to induce antibody-mediated immune responses. The triggering of an increased lymphocyte-mediated response has been the main strategy against TB during the past 20 years, although the results have shown that this may not be sufficient in terms of efficacy.34,35 In this sense, and although controversial for many years, it has recently been shown that humoral responses may play a more important role than commonly thought.36,37 The most recent research has focused on identifying new Mtb antigens using new techniques, including in silico approaches, some of which are expressed “in situ”, or lipid antigens,38 and newer and better adjuvants, as well as improved viral vectors or nanoparticles to carry antigens and thus improve their delivery to antigen-presenting cells.39

The efforts of the European Community (EC), AERAS, TBVI and the Bill and Melinda Gates Foundation to boost the TB vaccine field scenario over the last 20 years have been remarkable. The resulting injection of a mixture of funds and trust, and fostering of the networking and collaboration among the key scientists in this field, has led to major advances in terms of our understanding of protective immunity, the most appropriate animal models to be used and TB vaccine research and development. Numerous candidates have been generated in the past 20 years, and the EC alone (through the framework programs FP5, FP6, FP7 and H2020) has funded TB vaccine-related research that has generated half the TB vaccine candidates in the current global pipeline.38 Moreover, significant efforts have been made to harmonize protocols, technological platforms, and a lot of know-how has been generated, including on gating, downselecting candidates and the use and refinement of experimental animal models.47–49 Despite this, more funds are constantly needed if these candidates are to be moved forward in the clinical development process and reach the market. The licensing of a vaccine requires proven efficacy in the target population and the cost of a phase III clinical trial is very high. As we have mentioned above, the experience with BCG shows that it is less effective in people who have already been exposed to Mtb, and approximately 90% of latently infected individuals will never develop TB, so this is probably because the infection itself confers a certain degree of protection.50 Moreover, it is likely that what matters most in terms of developing active disease is the infective dose, as has been shown in experimental animal models. This is supported by epidemiological findings as those at highest risk are the close contacts of TB patients, and the risk of reinfection is higher in highly endemic settings.51,52 As such, in terms of a prophylactic vaccine intended to prevent TB infection, this leaves us little room for improvement as any candidate should do better than the infection itself, which indeed is already remarkably good.50

Moreover, considering all ages, forms of disease and the different distributions of TB endemicity and prevalence of comorbidities, it is highly unlikely that a single vaccine could be used for everything.38 In this context, mathematical modeling could help to identify which populations may benefit the most from a TB vaccine and should be included in the development of future TB vaccines.53

A vaccine that is able to reduce the risk of pulmonary TB (PoD) is needed as this would be a breakthrough in terms of stopping transmission. This could be achieved using a pre- or post-exposure vaccine that can decrease the risk of progression toward active TB. However, achieving an appropriate sample size to demonstrate efficacy in these conditions would imply large and lengthy clinical trials, which represents a high cost. As no definitive correlate of protection exists, in order to reduce the size, duration and cost of such clinical trials while predicting efficacy or identifying the individuals at high risk of developing TB, clinical endpoints are needed. Focusing on specific populations can help to reduce the sample size while demonstrating the biological activity of vaccine candidates.22 An assessment of the prevention of Recurrence (PoR) during the first year post-treatment (when most recurrences occur) can act as a surrogate endpoint, although the results will need further confirmation in larger trials for a vaccine to be licensed.

Although M. vaccae, RUTI and DAR-901 have been in the pipeline for a long time, therapeutic vaccines have only come into the spotlight very recently.26,54 The aim of therapeutic vaccines is to increase the cure rate of treatment regiments and/or to reduce recurrence rates. There are well-established WHO definitions for these concepts that imply evaluation at the end of chemotherapy (6 months for DS-TB and up to 2 years for MDR/XDR-TB).55 However, as the cure rates in drug-sensitive TB cases are very high at the end of treatment, the room for improvement is limited. In MDR and XDR-TB cases, in contrast, the recruiting rate can be even longer, thereby increasing the cost and hampering the feasibility. In order to mitigate this problem, month 2 sputum conversion rate for DS-TB and month 6 for MDR/XDR-TB should be considered as endpoints in all trials as both have been considered to correlate well with a patient's final outcome.56,57 Any improvement in terms of time to stable negativization of the sputum would provide a reasonable indication for using the vaccine to shorten treatment and could have a marked impact on controlling the spread of the disease.58 However, it is important to stress that the rate of negative outcomes in MDR/XDR-TB is so high that is worthwhile to design new strategies involving a vaccine administered as coadjuvant to chemotherapy.26,59

In summary, after 20 years of intensive research we still do not have a vaccine that is able to replace BCG and there are still many unknowns in this field.35,51 However, our understanding of TB vaccinology has been substantially expanded and there are up to 17 candidates in the global pipeline, with several more yet to enter clinical development.22,49 We are confident that if funding levels are maintained, the scientific community will be able to harvest the results of all the work done over the past few years and that a brighter future lies ahead.

PJC was the inventor of RUTI and one of the founders of Archivel Farma, the company that developed the vaccine, and was its CSO until 2013. CVM worked on the preclinical and clinical development of the RUTI vaccine.