India has successfully developed its first-ever truly indigenous vaccine, Rotavac, against the organism rotavirus that is responsible for many cases of diarrhoea. However, the Indian medical community is divided over the issue of including vaccines against rotavirus in the UIP. By R. RAMACHANDRAN

ON May 14, following the completion of Phase III clinical trials, the Department of Biotechnology (DBT) and Bharat Biotech, a Hyderabad-based company, announced in New Delhi the successful development of a home-grown oral vaccine against the organism rotavirus that causes killer diarrhoeal disease in children under five years of age. The data of the trials were presented at the International Symposium on “Rotavirus Vaccines for India—the Evidence and Promise” hosted by the DBT, the main Indian agency that had backed the development of the vaccine through all the stages under the Indo-U.S. Vaccine Action Programme (VAP).

This marked the culmination of nearly three decades of effort, including research by a consortium of Indian and U.S. scientists, creation of production capacity involving an Indian industrial partner, multi-centric vaccine trials in three phases, and the final stage of launching a viable vaccine for nationwide use. The effort is aimed at bringing to the market an indigenous, safe, efficacious and affordable vaccine against rotavirus.

This will indeed be the first-ever truly Indian vaccine: an Indian strain, an Indian manufacturer, clinical trial data in the Indian context and funding by the Indian government. “It is a unique social innovation model involving experts from a range of agencies from both countries and with costs of development shared by several partners,” said M.K. Bhan, the former Secretary of the DBT whose research at the All India Institute of Medical Sciences (AIIMS) in the mid-1980s was key to the development of the vaccine.

India today accounts for nearly 22 per cent of the world’s mortality burden from diarrhoeal disease, which translates to about 100,000 children’s lives every year. Equivalently, roughly one child in 242 dies from a rotavirus infection before five years of age. The death burden is highest among the youngest—about 50 per cent and 75 per cent of deaths occur before one and two years of age respectively (Figure 1 and Table 1). A recent Global Enterics Multicentre Study (GEMS), which was a case-control study of acute diarrhoea in children under five conducted at seven sites in Africa and Asia, including the National Centre for Cholera and Enteric Diseases (NICED), Kolkata, has found that rotavirus is responsible for the highest number of diarrhoea cases at the Indian site.

Bharat Biotech has successfully established an indigenous vaccine production capacity to cater to the Indian market, but the vaccine, called Rotavac, still has to go through the approval and licensing process of the Drug Controller General of India (DCGI). According to Krishna Ella, the chief executive officer of the company, the application for licence will be made by July.

Two foreign vaccines have already been licensed for marketing in India: Rotarix of GlaxoSmithKline and RotaTeq of Merck. They have already been introduced in more than 40 countries. In fact, on the basis of their performance in trials in the developing-country settings of Nicaragua, Mexico and Brazil, the World Health Organisation (WHO) has recommended their introduction in national immunisation programmes in Africa and Asia.

While researchers and Bharat Biotech are positioning Rotavac for inclusion as part of routine immunisation under India’s Universal Immunisation Programme (UIP), its introduction will depend on the evaluation and recommendation by the National Technical Advisory Group on Immunisation (NTAGI). The company has targeted a price of about Rs.50 a dose of Rotavac for the national programme and a somewhat higher price for the private sector. Rotarix and RotaTeq each cost over Rs.1,000 in India. The inclusion of Rotavac in the UIP may not be straightforward as not everyone in the Indian medical community is convinced that this is the right thing to do.

WHO’s seven-point plan

Vaccination is a part of a seven-point action plan recommended by the WHO and UNICEF (United Natons Children’s Fund) to combat childhood diarrhoea. The treatment component of it comprises (i) fluid replacement to prevent dehydration and (ii) zinc supplementation and the prevention component comprises (iii) rotavirus and measles vaccinations, (iv) promotion of early and exclusive breastfeeding and vitamin A supplementation, (v) promotion of hand washing with soap, (vi) improved water supply in quantity and quality, including treatment and safe storage of household water, and (vii) promotion of community-wide sanitation.

Vaccination as sole strategy?

One of the main arguments against vaccination as the sole strategy to combat diarrhoeal disease is that there are substantial differences in disease burden between high-income, middle-income and low-income countries. This is ample evidence of the role that economic development, better sanitation, safe water supplies and better health systems play in reducing morbidity caused by diarrhoeal diseases, and there are systematic studies to support this. Even within India, rates of mortality caused by diarrhoeal illnesses are higher in lower-income States than in higher-income States, suggesting that the rates can be reduced substantially even without rotavirus vaccination by providing access to safe water supplies and better sanitation and health care. Moreover, in up to 58 per cent of cases that test positive for rotavirus, there is co-infection with other pathogens. In such a scenario, it is argued, it would be inappropriate to assign excessively greater weightage to rotavirus vaccination as the primary strategy for the prevention of rotavirus diarrhoea.

Data from Mexico showed that two successive, naturally occurring rotavirus infections were found to protect completely against subsequent infections. But this, it is argued, cannot be used as evidence that any rotavirus vaccine will provide similar protection against all rotaviruses in all parts of the world and is another major argument against vaccination as the primary strategy against prevention of rotavirus. There are numerous strains in India that are different from those in other regions, and new strains are continuously emerging through reassortment between animal and human strains. Studies in India, it is pointed out, had found little evidence of such natural protection in the Indian context perhaps because of continuously evolving strains.

Gagandeep Kang of the Christian Medical College (CMC), Vellore, who was part of one study and was associated with the Phase III clinical trial for Rotovac, points out that this observation is not quite correct. “It does happen here as well but it happens a bit slower,” she says. The protection was only 79 per cent after three infections as against 100 per cent in Mexico after the first two infections. In fact, commenting on the study, the Nobel laureate Robert Baltimore had said: “The high reinfection rate may be because primary infections occurred very early—before the immune system had matured—and suggests that in India and similar settings vaccine efficacy might be enhanced by increasing the dose or number of doses or by vaccinating earlier.”

“We think,” Gagandeep Kang says, “that this happens with all oral vaccines in the Indian context; it’s not rotavirus alone. The same thing holds for polio, oral typhoid vaccine and oral cholera vaccine. There’s a laundry list of reasons as to why there is interference—multiple infections, mother’s breastfeeding with high antibody levels [which can neutralise the vaccine virus]. It may be some sugars. There’s a lot of interest in small sugar molecules in breast milk and how they can block [vaccine] viruses [from attaching to the gut and replicating].”

Another argument against the WHO recommendation that rotavirus vaccination be included in national immunisation programmes is that it is based on an extrapolation of efficacy data from one population to other populations that are in an “equivalent child mortality strata”, which may not be valid. Critics argue that there is not a single study that has looked at the efficacy in India of the licensed foreign vaccines, while studies in Bangladesh and Vietnam with RotaTeq in 2009 showed the vaccine efficacy to be only 48 per cent. Further, the results of a multi-centric surveillance in India between 2005 and 2007 found that Rotarix covered only 22.1 per cent of the strains identified in the study and RotaTeq only 47.9 per cent. Thus, evidence of cross-protection in one region does not imply cross-protection in India.

The Phase III clinical trial of the Indian rotavirus vaccine, which is based on an indigenous asymptomatic, or non-disease-causing, strain called 116E, should, however, have an important bearing on the ongoing debate on the introduction of a rotavirus vaccine as part of the UIP because its results have thrown up interesting science. Bhan and his colleagues identified the strain in 1985-86 from a newborn in the neonatal unit of the AIIMS. Around the same time, a group at the Indian Institute of Science (IISc), Bangalore, led by Durga Rao had identified a similar asymptomatic strain, called I321, from newborns in hospitals in Bangalore and Mysore.

Both these unusual strains were human-bovine gene reassortants, which means they include genetic elements from both human and bovine rotavirus strains. Since these were asymptomatic, it implied that the disease-causing components of a human strain had been switched with the disease-causing components of a bovine strain. That is, the strains were naturally attenuated. Both were taken as potential vaccine candidates as infants naturally infected with these strains did not, on the one hand, develop the disease and, on the other, demonstrated strong protection against rotavirus infection and did not develop severe diarrhoea when they were exposed to the pathogen again.

Studies and trials

Over two decades, a consortium of Indian and American scientists collaborated towards the vaccine development under the aegis of the Indo-U.S. VAP, which was initiated in 1987 between the DBT and the U.S. National Institutes of Health (NIH). The consortium included scientists and health experts from the DBT; the Indian Council of Medical Research; the IISc; the AIIMS; the National Institute of Immunology (NII), New Delhi; the CMC; the Society for Applied Studies (SAS), a non-profit private society engaged in health research; Bharat Biotech; Stanford University School of Medicine; the NIH; the U.S. Centres for Disease Control and Prevention (CDC); and the Programme for Appropriate Technology in Health (PATH), a non-profit international non-governmental organisation.

Two independent groups began working in parallel with the strains 116E and I321 respectively. In 1998, Bharat Biotech was identified as the partner for the next stage of vaccine development. The first Indian trial began in May 2003 with both candidate vaccines at the AIIMS, and the trial was completed in July 2003 with no major adverse events. Two Phase I clinical trials in adults and children (2-12 years) with rotavirus antibodies were first carried out at the Vaccine Trial and Evaluation Unit at the Cincinnati Children’s Hospital, U.S., and the vaccines were found to be safe.

Phase I studies in Indian children and infants were completed between May 2004 and May 2005 using NIH-produced vaccine lots. The results showed that both vaccines had comparable safety and viral shedding (indication of replication of the vaccine virus in the gut) profiles, but the 116E strain was found to confer greater immunity to disease—36.6 per cent of 116E recipients showed immune response attributable to the vaccine (measured in terms of serum antibody titre) as against 15.4 per cent of I321 recipients. This led to the dropping of I321 as a potential vaccine strain.

Also, studies at the CMC had found that I321 was not entirely asymptomatic in neonates and did cause diarrhoea. “The reason that we worried about the strain I321 being taken forward into a vaccine was because we saw it was associated with diarrhoea,” says Gagandeep Kang. “That, however, never happened with 116E. We have seen thousands and thousands of stool specimens. We have never seen 116E associated with diarrhoea. I have seen I321, largely in neonates and a little bit in slightly older children, never a 116E.”

The next trial (Phase IIa), a study using higher doses with the vaccine manufactured by Bharat Biotech based on starting material from the NIH, was carried out between November 2006 and February 2008 by the SAS. No safety concerns showed up, and a robust immune response was seen in 89 per cent of the infants after the third dose of the vaccine at high dose levels. In April 2008, on the basis of the encouraging Phase II trial results, Bharat Biotech proposed Phase III trials (efficacy study). The proposal, which was approved by the consortium, was to use the vaccine produced in the company’s own Vero cell-based production facility instead of vaccine lots produced at the NIH. (Vero cells are host cells used in cell cultures to grow viral stocks for vaccine production. A Vero cell line is derived from the kidney cells of an African green monkey and is a continuous cell lineage that can be replicated through many cycles of division without degradation of the culture.) In addition to resources provided by the DBT and the company itself, funds from the Bill & Melinda Gates Foundation, the Research Council of Norway and the Department for International Development (DFID) of the United Kingdom were routed through PATH.

The all-important Phase III trials began in March 2011 with 6,799 infants enrolled from across three sites in the country: 3,799 at the Centre for Health Research and Development of the SAS in New Delhi, 1,500 at the Shirdi Sai Baba Rural Hospital at the KEM Hospital Research Centre in Pune and 1,500 at the CMC. Nita Bhandari of the SAS led the trials. Of the 6,799 infants in the randomised, double-blind, placebo-controlled trials, at the time of the first dose 4,532 infants were administered Rotavac and the remaining 2,267 a placebo. Rotavac was administered as a liquid formulation stored at –20 °C.

Excellent safety profile

Three doses of the vaccine were given: at six to seven weeks old (at the time of enrolment), at more than 10 weeks, and at more than 14 weeks of age, with 8 months being the maximum age up to which the Rotavac/placebo was administered. The total number at the time of the third dose was 6,546 (4,356 + 2,190), implying over 96 per cent compliance. Viral shedding in stool specimens was evident of vaccine virus replicating, and this was seen from three to seven days after the first dose.

An independent expert group, headed by Mathuram Santhosham of Johns Hopkins University, served as the Data Safety Monitoring Board (DSMB) for the Phase III trial. According to the DBT, on February 13, the DSMB certified that the trial complied with international standards for good clinical practices and stated that the safety profile of the vaccine was excellent and that the efficacy data obtained from the trials met pre-specified criteria for success. It is also supposed to have given its nod to move towards securing licence for marketing at the earliest.

“The clinical study,” stated the DBT press release, “demonstrates that… Rotavac is efficacious in preventing severe rotavirus diarrhoea in low resource settings in India. Rotavac significantly reduced severe rotavirus diarrhoea by more than half —56 per cent during the first year of life, with protection continuing into the second year of life.

Moreover, the vaccine also showed impact against severe diarrhoea of any cause.” But, most importantly, Rotavac seems to confer broad protection against most commonly circulating genotypes of rotavirus, with vaccine efficacy ranging from 31.3 per cent against the strain G1P8 to 69.9 per cent against the strain G12P8 (see Table 2) though it may be noted that G1P8 is one of the prevalent strains in the country.

“It is the host factors that are largely responsible for vaccine efficacy,” said Bhan. “The efficacy of all the vaccines will be about the same in the Indian context and the performance of all vaccines will be roughly similar in similar environments.” On the question of cross-protection, he said, “Even though they are based on different strains, cross-protection is not an issue and the data clearly show that. The genotype of the vaccine strain seems to be irrelevant,” he added. While 116E is an asymptomatic strain, the strains of both Rotarix and RotaTeq are based on disease-causing strains. “The edge that Rotavac has over other vaccines, being naturally attenuated, is only in terms of safety and cost,” Bhan said.

The architecture of rotaviruses is quite complex, and each genotype is classified as a G-type and P-type. The outer shell of the virus (capsid) consists of three concentric protein layers (Figure 2). The outer layer consists of the viral proteins VP7 and VP4, the latter sitting as spikes on the surface. VP7 is a glycoprotein and determines the G-type of the virus and VP4 determines the P-type of the virus and also how virulent the strain is. The intermediate layer consists of the protein VP6, which determines the group of the virus (and here we are concerned with Group A rotaviruses). The proteins VP6 and VP2 together with the enzymes VP1 and VP3 make up the viral core. The capsid surrounds the viral genome, which comprises 11 free bits of double-stranded RNA that encode for six non-structural proteins (NSPs) and six structural proteins (VPs).

Rotarix is an attenuated monovalent vaccine and has only the human strain G1P8 in it. RotaTeq, on the other hand, is a pentavalent (combination of five strains) vaccine and that too a human-bovine reassortant vaccine, with reassortment done in a cell culture during vaccine development. When a human strain and an animal strain, say bovine or porcine, are mixed, the 11 genes will reassort and produce different combinations of these genes. So from a parent human and animal strain, one ends up with different reassortant strains.

The background of Rotateq is a bovine G6P5 strain. Of the five reassortant strains of RotaTeq, four are reassorted for G and the fifth reassorted for P. The Rotavac strain, 116E, is a natural bovine-human reassortant —now adapted as a human strain —with human G9 and bovine P11 and all other 10 segments human. The I321 strain of the IISc is also a bovine-human reassortant and has nine bovine segments and only two segments of human origin.

“There are several new features of the 116E strain,” points out Gagandeep Kang. “It is a natural reassortant that has never been known to cause symptomatic disease. This is unlike any other human virus that has been developed into a vaccine. It is a neonatal strain. We did not know whether it would replicate in older children and produce sufficient immune response to afford protection. Now we know that it can,” she adds.

Immunity mechanism

According to scientists, though several rotavirus vaccines have been developed, a full understanding of the mechanism of conferring immunity against natural rotavirus infection is lacking. The outer capsid viral proteins VP4 and VP7, which determine the virus genotype, are responsible for virus attachment and entry into susceptible cells. For vaccine development it was generally believed that antibodies against VP4 or VP7 could block virus growth and replication in a type-specific manner—homotypic protection—and perhaps in a limited cross-protective—heterotypic—manner. But now the empirical evidence that all rotavirus vaccines have about the same efficacy and that they all confer cross-protection seems to suggest this may not really be the mechanism underlying immunity against rotavirus.

“Serum antibodies,” says Gagandeep Kang, “can be neutralising and non-neutralising. Neutralising antibodies will prevent the virus from going into cell. The only neutralising antibodies that you have are directed against VP7 and VP4. The reason why people thought that VP7 and VP4 were going to be very important and were going to give type-specific immunity was because these are the only two known antigens inducing neutralising antibodies. But in vaccine trials, when you measure immunogenicity, you actually measure the non-neutralising antibodies that are directed against VP6. So are these relevant for protection? We don’t know. At least from the vaccine trials these seem relevant.

Earlier we thought neutralising antibodies were going to be critical and that’s why everyone was making multivalent vaccines. What the trial data is showing is that although we do not know what’s important for protection, this is clearly not the case.” At least, neutralising antibodies do not play a dominant role in protection against rotavirus infection. Since non-neutralising antibodies do not recognise VP4 and VP7, the protection they confer is the same across all genotypes.

In fact, as far back as 1996 Harry Greenberg of Stanford School of Medicine, who is part of the consortium for the development of Rotavac, and others had observed that the non-neutralising antibodies in mouse models had a protective role. Later, in 2002, Greenberg and colleagues proposed a mechanism to explain how non-neutralising antibodies against VP6 might inhibit rotavirus replication. Perhaps, that is the central mechanism for immunity against rotavirus. “I work with Harry, and we have just got a new grant from the NIH and the DBT to address the issue of which antibodies are generated in infection and vaccination with the first and the second doses. We hope that the results of the pilot [study] will take us to the next step of determining, based on immune response, can we predict who will and will not be protected? There are years of work ahead of us,” says Gagandeep Kang.

Notwithstanding the controversy over the introduction of a rotavirus vaccine in the UIP, and irrespective of whether Rotavac will be approved for use in the national programme, the trial results, soon to be published in The New England Journal of Medicine, have provided new insights into the evolving science of immunity against rotavirus.