FAILING THE COVID-19 TEST
The furore over Chinese rapid antibody testing kits reveals a number of problems, from ‘faulty’ equipment, long supply chains, disagreements on interpretation and even crisis-profiteering
The low accuracy of the Chinesemade rapid antibody testing kits has hurt the containment strategy in many states. Where did we go wrong?
OOn April 16, a special Air India cargo flight landed in New Delhi, carrying a consignment of 276,000 rapid antibody testing kits from Guangzhou Wondfo Biotech, a Chinese medical technology firm. Designed to test for possible exposure to COVID-19 by detecting the presence of antibodies in blood samples, the arrival of these kits was greeted with much enthusiasm. They were expected to play an important role in India’s efforts to map the spread of the virus.
India, which had tested 770,764 samples as on April 29, is still far from reaching the testing rates of countries like South Korea or the US, which have preferred mass testing over lockdowns. Both those countries have made wide use of rapid antibody tests, one of the reasons their testing numbers are so high.
Suresh Kulandaivel, director of Matrix Labs, Chennai, which holds a licence from the Drugs Controller General of India (DCGI) to import the testing kits, had been awaiting the shipment for almost a week. “There were some delays in custom clearances,” he says. His relief at their arrival was to be short-lived. Once received, the Indian Council of Medical Research (ICMR) in Delhi distributed them to states across the country; but soon, complaints started pouring in from Rajasthan, Maharashtra and West Bengal, disputing the accuracy of the results these kits provided.
Rajasthan, which received 30,000 kits, used them to test a number of samples, of which 36 were found to be positive. However, follow-up confirmatory tests using the PCR (polymerase chain reaction) method turned out negative. To clear up the matter, Rajasthan additional chief secretary of health Rohit K. Singh set up a committee of doctors and microbiologists to investigate. They administered the antibody test to 168 patients who had tested positive for COVID-19 via the PCR method at least eight days prior—long enough for antibodies to have appeared in their blood samples. The Chinese test kits identified only ninc of them as COVID-19 positive. A notice was then sent to the ICMR, informing the agency that the kits had an accuracy of just 5.4 per cent.
QUALITY CONTROL
On March 27-28, the ICMR had cleared a tender for 500,000 kits from Wondfo, with another 250,000 kits being ordered from another Chinese firm, Zhuhai Livzon Diagnostics. The agency says these kits were being procured at the request of states. On April 5, after the ICMR had placed its order, researchers at Oxford University found that a batch of Wondfo kits, which the UK had placed an order for, did not perform well or meet the criteria for a reliable test. The UK reportedly paid $20 million for around 2 million kits and is now hoping for a refund. “If we were procuring kits from China and questions about their quality were being raised globally, more due diligence should have been done,” says Malini Aisola, co-convenor of the All India Drug Action Network (AIDAN).
However, the tender for the 500,000 kits from Wondfo did include a key clause which stated that the ICMR reserved the right to do random-lot testing or pre-shipment testing for quality. “I was surprised when I heard [that countries are] depending so heavily on antibody testing,” says Leena Menghaney, regional head for Médecins Sans Frontières, who has worked closely on HIV and viral hepatitis diagnosis in India. “In the past, [when using antibody kits to test for] viral hepatitis, we found that judging the quality of such kits is difficult and that they have had a history of high false negatives.”
According to the terms of the ICMR tender, any item found not acceptable would be returned and was to be replaced in a timely manner. Whether the agency performed random-lot tests on the 276,000-kit consignment when it was received on April 16 is uncertain. The ICMR did not respond to enquiries on this point.
WHERE DID WE GO WRONG?
On April 27, the ICMR issued a notice asking states to halt the use of these testing kits and for them to be returned. The ICMR’s notification stated that the kits had ‘shown wide variation in their sensitivity, despite early promise of good performance’. This led to a rebuke from the Chinese embassy in India. Chinese counselor Ji Rong called it “unfair and irresponsible for certain individuals to label Chinese products as faulty”. She, and both the manufacturers, have stated that the kits are certified, suggesting that improper storage or handling might be the problem. Given the lack of an official enquiry so far, it is difficult to assign blame; but there are some potential mistakes already visible.
First, we may have misjudged quality from the start. International bodies like the European Commission have laid down some essential parameters on which rapid antibody kits should be judged. This includes the type of test, the nature of the test, the measured target, individuals that these tests can be used on, the type of specimen required, how the tests are to be stored, the required qualification for staff using them, guidance on interpretation of results and the potential limitations in terms of false positives and negatives.
“I sent a sample kit to the National Institute of Virology, Pune (NIV-Pune), for testing along with all relevant paperwork, before I was issued a licence,” says Kulandaivel. The protocol NIV-Pune used to approve the Wondfo kits has not been made public. It is also not clear if the other rapid kit brands approved by the laboratory were graded in terms of quality, ease of use and other potential limitations, and if these formed the basis on which the ICMR decided to order from Wondfo.
The use of these kits and interpretation of their results, as pointed out by the manufacturer, could have also been an issue. Not all states have reported
THE ICMR HAS ASKED STATES TO STOP USING THE KITS IT HAD PROCURED, SAYING THEY WERE FOUND TO BE UNDERPERFORMING
problems with the kits. Chhattisgarh and Gujarat, for example, have not done so. Gujarat even went ahead and began using the 24,000 kits it received from the ICMR in 30 of its 33 districts. Gujarat principal secretary of health, Jayanti Ravi, says the kits were calibrated for use and training was given to healthcare workers on how to use them. “The test is only for surveillance, to give us an idea of which areas might have exposure to COVID,” says Ravi.
Chhattisgarh, which received 4,800 kits, had conducted 392 tests, of which four were positive. “We didn’t have any issues with the [Chinese] kits. But given the questions about their accuracy, we decided to place an order with a South Korean firm, SD Biosensor, for 25,000 kits at Rs 400 each. These have arrived. We will begin using them this week, but only to check immunological status and see if there is a possibility that the virus could be present in a hotspot or non-hotspot. If a person is positive in a rapid test but negative in PCR, it could suggest they were a carrier at some point. Such data is useful for contact tracing,” says Chhattisgarh health minister T.S. Singh Deo. SD Biosensor has said it has also received an order for 100,000 kits from the ICMR. Its tests are manufactured at a plant near Manesar, Haryana.
The Chinese kits have specific usage protocols. “From what I have seen of the rapid test protocols, most of them need to be stored at a certain temperature—around 25-30 degrees Celsius,” says Dr Arjun Dang, a pathologist and CEO of Dr Dang’s Lab, one of India’s leading pathology labs. “If the kits are left out [in the open] for even a short amount of time, it could affect their accuracy. After being opened, they must also be used within a specified time period.”
Wondfo Biotech has agreed to cooperate with Indian agencies looking into the matter. In a letter, the company also pointed out that detection rates for antibody tests become higher as the body recovers from the virus, and that those who have tested positive with an antibody test might show up as negative in a PCR test if the virus has already been neutralised by the antibodies. Similarly, a negative result in an antibody test could appear alongside a positive result in a PCR test because the virus is present but antibodies have not built up in the blood. “One test detects the virus, the other detects the antibodies,” says Dr Sreedhar Mayavan, managing director of PrimeGen Healthcare Labs in Chennai. “Both might not be present at the same time. Antibody tests are best used to detect past exposure to infection. Different tests are used at different stages of viral infection. Moreover, to evaluate any diagnostic kit, you need a greater range of samples than a few hundred.”
PRICING PROBLEMS
There are other issues attached to the rapid kits controversy. A major allegation is that of private profiteering.
Chinese companies expect advance payment for most medical equipment. For the Wondfo consignment, Matrix Labs paid the full amount up front—Rs 12.25 crore for 500,000 kits. However, Matrix did not sell these kits directly to the ICMR. That sale took place through a dealer, Delhi-based Rare Metabolics, which claims to be the sole distributor for Matrix’s imported kits. (Though Matrix sold kits to the Tamil Nadu government through a different dealer, Shan Biotech & Diagnostics.) Rare purchased the kits from Matrix at Rs 21 crore, and then sold them on to the ICMR for Rs 30 crore, bringing the price to Rs 600 per kit. Matrix says that while Rare promised full payment, it has only received Rs 12.75 crore so far. A petition was filed in the Delhi High Court on April 24 to settle the matter. The final understanding was that Matrix would supply the remaining 224,000 kits due to the ICMR and that Rare would pay Matrix the full amount of Rs 21 crore for 500,000 kits.
On April 27, Matrix received the remaining 224,000 kits, as well as a fresh consignment of 500,000 more kits. (Of this new batch, 50,000 kits will go to the government of Tamil Nadu, which had placed a separate order at the same rate as the ICMR.) This leaves Matrix with 450,000 kits, which the high court said it could sell, at no more than Rs 400 per kit, directly to any state government or private entity it wished to. Kripa Shanker Gupta, director of Rare Metabolics, declined to comment when contacted.
The court did not rule on whether the ICMR needs to pay Rare Metabolics the full Rs 30 crore for the 500,000 kits. The ICMR has clarified that the price band at which it was permitted to procure rapid tests was Rs 528-795. Four bids were received for the tender issued and it picked the lowest price quoted. “The raw cost of rapid kits is minimal,” says Aisola. “Why we decided to have such a long supply chain instead of procuring kits directly from the importer to keep costs at the minimum is a mystery. If it was done to keep costs low, the quality has turned out to be poor. If it was done in the interest of time, it has not served its purpose. We are in the fifth week of lockdown and wide-scale testing in all states is yet to take off.”
EXPERTS ASK WHY THERE IS SUCH A LONG SUPPLY CHAIN, SAYING THAT KITS SHOULD BE PROCURED DIRECTLY FROM IMPORTERS, RATHER THAN THROUGH DEALERS
NOW WHAT?
India’s testing rate varies significantly from state to state. Those with lower numbers, like West Bengal or Bihar (testing at 121 and 162 per million respectively), say they are awaiting supplies. States with high testing rates but also high rates of infection, such as Delhi and Rajasthan (testing at 1,880 and 1,885 per million, respectively), also say they need more kits. “From a quality and treatment point of view, the gold standard is a PCR test. India has a fairly good capacity to conduct PCR tests given our history with HIV, viral hepatitis and TB diagnosis,” says Menghaney. This is true. Not only have we cleared five companies to manufacture PCR kits in India but we have the existing infrastructure for it. Andhra Pradesh, for example, has a True-Nat machine—which uses PCR technology—in every public health centre in the state. However, the issue here is not a shortage of devices, but of the chemicals they use.
“The [production of the] reagents needed to convert RNA to DNA in the PCR test [rests] largely with Roche and Thermo-Fisher,” says Menghaney. There is currently a global shortage of these chemicals, with France, the US and the Netherlands reporting diminished reserves. There is no reported data as to the Indian stockpile of these critical resources.
Given the shortage of reagents and our large population, rapid kits are necessary for our national cluster containment strategy, helping public health officials quickly identify and segregate areas which could have had some exposure to COVID-19. The DCGI and ICMR have approved 100 suppliers for rapid kits. Of these, 68 firms supply Chinese kits and 14 of them sell kits from Wondfo. While states are free to order from any of these suppliers, they will need to invest heavily in quality control.
As the country prepares to come out of lockdown, the only sure way to identify asymptomatic COVID-19 spreaders will be reliable diagnostic tools. The ICMR has clarified that the faulty Chinese kits have not cost the country a single rupee. However, it has cost us time and confidence, something we can ill-afford in a pandemic. ■
—with Romita Datta, Kiran D. Tare and Rohit Parihar