Just over a week ago, Dr. Charles Chiu analyzed the latest batch of genomic sequencing results from his laboratory at UCSF. He was examining the family pattern of a particularly infectious coronavirus variant in the UK, known as B117.
He didn’t find it. What he discovered, instead, was alarming and unexpected: a rare variant that Chiu had seen only a few times, suddenly represented 25% of his samples. California had a mysterious new enemy to contend with.
“It slipped under our noses,” said Chiu, who is director of the UCSF-Abbott Viral Diagnosis and Discovery Center.
Chiu’s team was the first to report the new variant, which the state is calling L452R, but days later another group of scientists from Cedars-Sinai reported that the same variant was now responsible for more than a third of the cases in Los Angeles .
The findings underscore the urgent need to increase genomic sequencing in California and the country, say scientists like Chiu. California has more laboratories capable of doing this job than many countries, but the state lacks the coordinated infrastructure needed to quickly identify new variants and determine whether they are a threat.
The same is true for the United States, which is far behind almost every other large country in terms of sequencing, and lacks funding and national leadership to improve the situation, say infectious disease experts. As the nation races to vaccinate as many people as possible before the virus catches more mutations that might make it more difficult to control, it is critical that scientists understand what they are up against.
Without much more large-scale surveillance, said Chiu, everyone who tries to control the pandemic is “flying blind.”
“We don’t know which variants may be circulating or emerging. It is very difficult for us to fight an enemy if we don’t know what he is, ”said Chiu.
Genomic sequencing is used to determine the order of the chemical building blocks that make up the virus. These building blocks change, or mutate, as the virus replicates. Genomic sequencing allows scientists to identify these mutations, providing a kind of genetic fingerprint for the virus.
It is important for several reasons. It helps scientists and public health officials to understand how the virus behaves at the molecular level. It can also help infectious disease researchers track where the virus is spreading and identify outbreaks. And it can alert authorities to mutations in the virus that can cause a change in behavior – making you more infectious or less susceptible to vaccines, for example.
This latter use is critical, but requires a robust national surveillance program that involves frequent genomic testing of random samples collected across the country. The United Kingdom has such a program that tests about 10% of coronavirus cases. The United States tests well less than 1% of its cases.
“I know what this country can do. But there has to be an appeal about that, and it has to come from above, ”said Dr. Joe DeRisi, co-president of Chan Zuckerberg Biohub in San Francisco, who does about 45% of genome sequencing in California and 5% of national total. “It can be done. The UK did it. The UK has wonderful genomic surveillance for the country. Why are we so late? No national leadership.”
California’s genomic sequencing network is poorly coordinated, with a handful of laboratories across the state providing results, but few guidelines for what types of samples are collected and how they are reported. Most laboratories that do genomic sequencing do not focus on surveillance, which involves regular screening for mutations in the virus, but on the analysis of specific cases and outbreaks.
And there is little or no funding, said DeRisi. Biohub and UCSF pay for genomic sequencing with concessions they have already obtained.
“We are trying to do what we can. We will donate as much effort as possible to this ”, said DeRisi. “But we are a small non-profit research institute and it looks like we are carrying a very heavy burden.”
In addition to UCSF and Biohub, UC Berkeley and Stanford also do the genomic sequencing of the coronavirus. Stanford said last week that he started a surveillance program to quickly find new variants that may be circulating in northern California.
Chiu’s laboratory found two of the first cases of variant B117 in California, which were reported in San Bernardino County. The Scripps survey in La Jolla identified a group of cases of B117 in San Diego County. Another case in Los Angeles County was found by a federal laboratory.
This scattered approach to finding new variants, both at the state and national levels, is not capturing how widespread they may already be, said Dr. Joel Ernst, an infectious disease specialist at UCSF. The B117 variant has been found in 22 states so far, including 72 cases across California, according to the Centers for Disease Control and Prevention.
“We know, for example, that the UK variant is here. We know it is in several states, ”he said. “But we have no idea what fraction of infections in Colorado, Los Angeles, Seattle or Boston are caused by this variant.”
It will be more important than ever to increase genomic sequencing as more people are vaccinated, infectious disease experts said. A small percentage of vaccinated people will develop COVID-19 anyway, and scientists need to be able to identify these cases and understand why the vaccine failed, said Ernst.
And as more people get vaccinated, the pressure will increase on the virus to mutate and try to escape. Public health experts will need to know quickly if a new variant does not respond as well to vaccines.
The B117 variant appears to be fully covered by current vaccines. But a variant identified in South Africa – which, like the UK, has a strong genomic sequencing network – may be somewhat resistant to vaccines, studies have found. There are similar concerns about a variant identified in Brazil.
“We are launching the largest evolutionary experiment in virology that has ever existed on the planet. We will put immunological pressure on the virus that will force it to mutate or die. There are likely to be mutations that can circumvent the vaccine, ”said DeRisi. “Wouldn’t we like to know that so we can design a generation 2 and 3 vaccine?”
Chiu’s laboratory has already started studies on how the L452R variant responds to vaccines. Three days after reporting the variant to the state, he started growing it in the laboratory. In about a week, he will expose the variant to antibodies collected from previously infected donors. This will tell him whether the virus has a mutation that allows it to escape one or more antibody weapons that the body uses to fight infection.
But the other important question is whether the variant is more infectious than other versions of the virus. “To show that it is more transmissible, a robust community surveillance program would be needed to see if the virus is increasing in frequency in relation to other mutations,” said Chiu. “This can only be done with a very structured, organized and coordinated surveillance program, which we don’t have.”
Chiu is concerned that the L452R variant may actually be more infectious. His team found that it represented less than 5% of the samples tested from November to mid-December, but 25% in a second batch from mid-December to January. It is a worrying increase in a very short period of time. The variant was also associated with several large clusters in Santa Clara County, including an outbreak at Kaiser hospital in which more than 90 people were infected.
The variant also carries a mutation located in a precarious location of the virus that can make it more infectious.
The Cedars-Sinai group suggested that the variant may have helped fuel the recent worrying increase in cases in Los Angeles County. But scientists like Chiu say they are not sure. All evidence is circumstantial.
Regardless of the size of the threat that the L452R or any other variant poses, the arrival of multiple mutations at the same time should raise a red flag, said Fyodor Urnov, virologist at UC Berkeley’s Institute of Innovative Genomics.
“It’s a bit like a tremor at the San Andreas fault. Like a 3.5, ”said Urnov. “It’s not a big quake, but if we don’t start preparing for the big one, Mother Nature will say, ‘Didn’t I tell you? What part did you miss? ‘”
The San Francisco Chronicle writer, Michael Williams, contributed to this story.
Erin Allday is a writer for the San Francisco Chronicle. Email: [email protected] Twitter: @erinallday