Why is India failing to use its genome sequencing capacity for Covid-19?

On May 25, Union health minister Harsh Vardhan while addressing the press said that 25,739 samples taken from Covid-infected people had been put through whole genome sequencing in India. This amounted to less than 0.1% of the total reported infections in the country at the time, woefully short of the target that the Centre had set in December: 5% of all positive samples.

Genome sequencing is crucial to detecting new variants of the coronavirus. Despite the massive shortfall in its efforts, India is not utilising existing sequencing capacity in the private sector. Virologist Shahid Jameel, who until last month headed the Indian SARS-CoV-2 Genomic Consortia, a network of 10 laboratories established by the health ministry for genome sequencing in December, said “there are a number of private labs with good capacity to sequence”.

But an order issued by the Indian Council of Medical Research in March 2020 expressly prohibits private labs doing Covid-19 molecular diagnostic testing from sequencing positive samples. Any violation, the order states, could result in “legal action”. The restriction was reiterated as part of an email communication ICMR sent to the private labs on April 26 this year.

“We got a communication very early on in the pandemic that we should not do anything with the positive samples,” said the head of a diagnostics lab in Hyderabad, who requested anonymity. The lab is capable of doing at least a couple of hundred sequences per day, he said. “I had recently written to them [ICMR] asking if we would be allowed to do so, and that we can help out in increasing the infrastructure. I was told that they were considering it.”

But the council’s spokesperson Rajni Kant Srivastava told Scroll.in that the restriction continued to remain in place and private labs should refrain from doing genome sequencing of Covid-19 samples. “Only the INSACOG should do the genome sequencing,” he said, referring to the consortia set up in December with 10 government labs. Eighteen more government labs were approved to be added to the consortia in May.

Allowing only select labs to do genome sequencing amounts to wasting resources, some scientists say. “There is sequencing capacity available in the country which is not being used and this is a mistake,” said Gautam Menon, who teaches physics and biology at Sonepat’s Ashoka University. “At this time, scientists would really like to know the spectrum of variants that exists in the country on a real-time basis.”

Priya Sampathkumar, infectious diseases expert at United States’ Mayo Clinic, concurred. “This is a mistake,” she said. “Increased capability to do sequencing is the only way to detect new variants.”

India and its lacklustre sequencing efforts

Sequencing the coronavirus genome and detecting new variants is important, scientists say, because it could provide us with a whole range of information that could help countries with their pandemic response. Among other things, it could tell us how the virus is mutating, possibly adapting to vaccines, and the way new variants are travelling across geographies. All of this is most effective when data is generated in bulk and fast and shared real-time.

But India has been a laggard when it comes to sequencing. Till December 2020, most of India’s genome sequencing were random efforts by individual scientists and researchers. The centralised Indian SARS-CoV-2 Genomic Consortia, or INSACOG, which was announced on December 30, started work only in mid-February. Not only did the process begin late, it has failed to scale up.

Consider this. On April 20, on the GISAID database, the largest global repository to report variants, the count of Delta sequences, belonging to the variant first found in India, was 660. Out of them, 298 were from India and 205 from the United Kingdom.

Given the variant’s high-threat perception, the United Kingdom upped its sequencing volumes. As of June 9, the country has submitted 23,015 Delta sequences to the database – nearly as much as the total number of Covid-19 samples India had sequenced till May 24. As for the Alpha variant, first reported in the UK, the country has submitted as many as 250,000 sequences to the global database.

The corresponding number of sequences submitted by India is 5,749, despite the country seeing millions of infections in April and May, many likely caused by the Delta variant.

One could argue that the GISAID database does not reflect the full picture since India does not submit all sequences there. But even accounting for that, the numbers are quite low. According to members of the INSACOG, India has so far sequenced “close to 40,000 samples”. That amounts to 0.13 % of total infections.

The number for the United Kingdom is 9.25 %. Closer home, Bangladesh and Sri Lanka have sequenced 0.19 % and 0.2 % of their positive samples. Even taking into consideration the fact the INSACOG started functioning only from February, India has sequenced only around 0.2 % of its reported cases since.

It is, therefore, “no surprise”, as epidemiologist Chandrakant Lahariya pointed out in The Hindu on June 9, that “we understand the Delta variant…far less than the Alpha variant”. Besides, much of what we know about the Delta variant and its relationship with vaccines is from research done in the United Kingdom, where the Covid-19 Genome Sequencing Consortium is a massive network comprising the four constituent nations’ public health agencies, universities and research institutions, the National Health Services’ regional laboratories and private diagnostic labs.

Given the clear need to generate more India-specific data, T Jacob John, one of the country’s foremost virologists, said he saw no reason why private labs which wanted to sequence should not be allowed to do so. “As long as the patient is not charged anything, I don’t see why it should not be done,” said John. “I thought the more we do, the more informed we will be.”

ICMR did not respond to Scroll.in’s queries seeking an explanation behind the rationale to bar private labs from doing sequencing.

Jameel, the former chair who quit INSACOG without citing a reason, said while he could not speak for the ICMR, he could think of two possible reasons: To “ensure that quality is maintained” and “one-off sequencing results are not released to the media without clinical and epidemiological correlations”. He said: “There is a certain protocol to be followed in terms of sample collection and analysis for it to make epidemiological sense.”

What Jameel was trying to convey is that it is important for sample collection and the subsequent analysis to be coherent because only then would the sequencing exercise provide desired insights into Covid-19 spread.

‘Smart sequencing’?

Scientists who are part of INSACOG say while adding private labs to the consortium was “very much under consideration”, adding more partners won’t add much value.

“Such discussions distract from the real problems,” said Anurag Agarwal, director of the Council of Scientific and Industrial Research-Institute of Genomics and Integrative Biology, one of India’s leading genome sequencing facilities. “What we need is smart sequencing, not surge sequencing.”

Agarwal said what he believed really mattered was “integration” and “analysis” of the data that was being generated – as opposed to the scale of the data. “Finding mutations is easy…the aim should be to find out how we can flag a particular mutant as dangerous before there is an outbreak,” he said. “If we really want to make a difference, we should reimagine this entire business including all the way going back to detection of variants at the time of diagnosis.”

Simply put, Agarwal is questioning the very rationale of conventional sequencing as the world knows it. He elaborated this idea in an article he recently wrote for Nature:

“Ideally, variant detection would happen in parallel with diagnosis, without sending samples to specialised sequencing labs. For that, we must develop simple variant tests that are affordable, easy for health-care workers to run using basic kit, and integrated into the systems that test, isolate and trace contacts. There are some promising new technologies, including low-cost CRISPR diagnostics for places with limited resources. Developing these should be prioritised.”

But many say much of what Agarwal is proposing is too far-fetched. “The technologies he is proposing are mythological,” said Jeremy P Kamil, associate professor of microbiology and immunology at Louisiana State University in the United States. “You’ll never see a ‘kit’ that can detect a variant on the spot… for detection of variants you need sequencing”.

“What they [Indian scientists] need to do is simple,” Kamil added. “Sequence more with faster turnaround time and upload to GISAID without delay so that all of the world can see the data.”

It was “disingenuous”, Kamil stressed, to suggest otherwise. “If we did not sequence we would not know what to focus on to limit vaccine escape – that is, only by sequencing can you know what variants might escape vaccines and then figure out how.”

Finding ‘appropriate’ samples

While Agarwal’s arguments of not expanding the network had more to do with the very utility of it, others also offered more logistical reasons.

Rakesh Mishra, who recently retired from the directorship of the Centre for Cellular and Molecular Biology in Hyderabad, one of the original 10 members of the consortia, said there were economic constraints to involving private labs. “Who will pay for it? Patient won’t,” he said. “And why would the government pay for it till its own institutes’ capacities are exhausted, which is not the case right now.”

The consortia, Mishra said, was not being able to sequence as much as it was meant to simply because it was not getting enough quality samples to sequence. “The bottleneck is not lab capacity,” he said.

While that may sound counterintuitive given India’s massive caseload, Mishra said the quality of data was contingent on “appropriate samples”. “Five percent randomised samples from all geographic locations in a country where people are tired and the system is overwhelmed is not easy,” he explained. In addition, the samples ought to be tagged with the correct metadata bearing details and clinical symptoms of the individuals associated with them.

Shouldn’t data alignment be easier for private diagnostic labs since they collect and process samples for molecular tests anyway? “We already get the form with the patients’ symptoms, etc,” said the founder of a chain of private labs based in Southern India. “So we could sequence a certain percentage of samples and there could be guidance [by INSACOG] on the number of symptomatic and asymptomatic we should do.”

Many of the bigger privately-owned networks of diagnostic laboratories in the country such as Strand Life Sciences and Metropolis have sequencing capabilities. So do some smaller stand-alone labs such as the Bengaluru-located Hybrinomics Life Science and Diagnostics. Prabhu Meganathan, who heads the lab, said they could “sequence “30-40 [sequences] per day” but “unless and until the government allows us to, we can’t”.

How much and who will pay?

As for the costs, the founder of the Southern India-based chain of private labs said the government could list out its requirements, float a tender and ask interested parties to bid. “Since this is not something that the labs are going to charge their consumers directly, it shouldn’t be a problem,” said the person.

What would the costs be like, though? It would depend on the scale, said Raja Mugasimangalam, founder and chief executive officer of Genotypic Technology, a Hyderabad-based company that specialises in sequencing and claims to have the capacity to do as many as 2,000 a day. As an example, Mugasimangalam said sequencing 100 samples would cost around Rs 8 lakh – that’s Rs 8,000 per sample. But for larger volumes, the per unit cost could be halved to Rs 4,000 per sequence.

While that may be considered steep, government labs which are part of the INSACOG are already outsourcing sequencing work to Genotypic Technology and other private companies. Mugasimangalam said his lab was currently sequencing a few hundred samples, significantly less than what it had the capacity to do.

‘Sequencing is not just about sequencing’

Mishra, the former director of the Centre for Cellular and Molecular Biology in Hyderabad, described it as a client-vendor arrangement where the private labs only do what they are told to. He seemed to suggest that the private labs were being employed to do the more mechanical part of the work while the INSCOG labs retained control of the actual analytical work.

“We give them the samples, they do the sequencing and submit the result to us over email,” said Mishra. But “sequencing is not just about sequencing and generating data”, he added. “You have to connect the dots and analyse it so that you can make contextual sense out of it”. Most of the Indian private labs do not have the capacity to do all of that, he claimed.

However, apart from the commercial laboratories, the teaching arms of several tertiary-care private hospitals that are treating Covid-19 patients have significant genome sequencing capabilities. Scroll.in spoke to two such institutes – Ramachandra Medical College in Chennai and Kasturba Hospital in Manipal – which said they could sequence samples if the government greenlighted it.

A professor at a third institute, Vellore’s Christian Medical College, confirmed that the clinical virology lab there had the infrastructure to do genome sequencing but was not doing it. However, the professor refused to confirm if it was because of the IMCR’s restrictions.

Considering these are full-fledged academic institutions attached to hospitals, it is unlikely that they would lack the competence to make epidemiological correlations.

Kamil said most of the arguments about keeping private hospitals out of the sequencing network were a “bit self-serving”. “If it is about making sure that data is shared, they should make sure that payments are not made unless the labs share data with the government consortium,” he said.

T Jacob John, who used to formerly head the clinical virology department in Vellore’s Christian Medical College, tended to agree. “I do not understand why ICMR is sensitive about genome sequencing,” he said. “In case any lab finds any new variant, it can insist that the original specimen must be kept safe and shared with INSACOG.”

The threat of legal action made even little sense, said Menon of Ashoka University. “These are legitimate scientific questions and the ICMR should have no monopoly on who gets to answer them,” he said. “This is in the interests of the country.”