Can re­searchers find a way to run thou­sands of COVID tests at once? Lo­cal team is try­ing


The goal: To run up­wards of 10,000 COVID-19 tests daily — and per­haps sig­nif­i­cantly more — by re­pur­pos­ing an ex­ist­ing “ul­tra-high-through­put” nextgen­er­a­tion genome se­quenc­ing plat­form. The tests would be in ad­di­tion to On­tario’s cur­rent lab­o­ra­tory ca­pac­ity. The team: Jeff Wrana, a se­nior in­ves­ti­ga­tor at Si­nai Health’s Lunen­feld-Ta­nen­baum Re­search In­sti­tute, is lead­ing the pro­ject along with co-in­ves­ti­ga­tors Ben Blen­cowe, Lau­rence Pel­letier, Rod Brem­ner, Tony Maz­zulli and Su­san Pouta­nen. The time­line: Wrana hopes to be ready to tran­si­tion to real-world test­ing in a month or two, but cau­tions that over­com­ing the lo­gis­ti­cal hur­dles out­side the lab will re­quire sub­stan­tial co-op­er­a­tion with other part­ners. The pro­ject was funded by the Univer­sity of Toronto’s COVID-19 Ac­tion Fund and the Krem­bil Foun­da­tion.

Be­fore the pan­demic, Jeff Wrana’s re­search had noth­ing to do with in­fec­tious dis­eases. He stud­ies can­cer, and has an in­ter­est in robotics. But when the COVID cri­sis hit, like so many other Cana­dian sci­en­tists, he won­dered what his lab could do to help.

Wrana and his col­lab­o­ra­tors turned to a suite of tech­nol­ogy used in his lab known as ul­tra-high-through­put next-gen­er­a­tion se­quenc­ing, which maps DNA in­cred­i­bly quickly. The plat­form is made up of two parts. The “next-gen­er­a­tion” part refers to ma­chines that can se­quence a sin­gle genome overnight by break­ing it up into mil­lions of tiny pieces and an­a­lyz­ing each one in­di­vid­u­ally. The “ul­tra-high-through­put” part refers to an ar­ray of robotics that au­to­mates the pro­cess­ing of huge vol­umes of sam­ples.

Plat­forms like this have had a ma­jor im­pact in the re­search world be­cause they give sci­en­tists the abil­ity to an­a­lyze thou­sands, or even hun­dreds of thou­sands, of bi­o­log­i­cal sam­ples at the same time.

Wrana, for ex­am­ple, has used it to an­a­lyze all the dif­fer­ent cell types that make up a tu­mour. By do­ing that, his lab can try to fig­ure out which are re­sis­tant to treat­ment and are at great­est risk of grow­ing back af­ter the rest of the tu­mour has dis­ap­peared.

“Part of the ge­n­e­sis of the idea was, well, if we can an­a­lyze 100,000 cells at the same time, then why can’t we an­a­lyze 100,000 pa­tient sam­ples at the same time?” Wrana says.

How would it work?

The way con­ven­tional test­ing for COVID-19 works is by search­ing for pieces of the virus’ ge­netic ma­te­rial in a pa­tient sam­ple — usu­ally a swab taken from the back of a per­son’s nose or throat. If the test finds it, the sam­ple is pos­i­tive.

This plat­form works the same way, but on a mas­sive scale: The goal is to take thou­sands or even tens of thou­sands of pa­tient sam­ples at one time, and let the tech­nol­ogy search for matches among all of them. Wrana and his team have gath­ered left­over swabs that have al­ready been tested for COVID -19 and run them through their se­quenc­ing plat­form to com­pare the re­sults.

“In a proof-of-con­cept co­hort, our per­for­mance is ac­tu­ally quite ex­cel­lent. It gets very good ac­cu­racy com­pared to the cur­rent clin­i­cal di­ag­nos­tic test,” he says.

There have been some snags. When they moved onto a larger batch of 1,000 sam­ples, they were thwarted, iron­i­cally, by On­tario’s suc­cess in con­tain­ing its epi­demic.

Be­cause the prov­ince has such low test-pos­i­tiv­ity rates right now — just 0.3 per cent of sam­ples test pos­i­tive some days, a very good sign — the feed­back “noise” from all the neg­a­tive sam­ples in­ter­feres with the pos­i­tive ones.

This prob­lem can be solved, how­ever. While none of this work is easy, re­pur­pos­ing the next gen­er­a­tion tech­nol­ogy for COVID test­ing isn’t the pro­ject’s big­gest chal­lenge, Wrana says.

What are the big­gest chal­lenges?

As the prov­ince has dis­cov­ered in its long and oc­ca­sion­ally painful jour­ney to in­creas­ing COVID test­ing ca­pac­ity, the path from swab to re­sult is a long one, with many op­por­tu­ni­ties for traf­fic jams.

Early on in the pan­demic, a ma­jor back­log of tests formed as de­mand over­whelmed ca­pac­ity. The lab net­work is cur­rently run­ning at high vol­umes: The prov­ince con­sis­tently pro­cesses more than 25,000 tests daily and some­times more than 30,000.

But the abil­ity to ex­pand even fur­ther, and even of­fer test­ing on a pop­u­la­tion scale — cy­cling through big chunks of On­tar­i­ans at reg­u­lar in­ter­vals — would still be help­ful, es­pe­cially for fre­quent screen­ing in high-risk set­tings and jobs.

As schools re­open, for ex­am­ple, the abil­ity to screen large num­bers of stu­dents and staff on a daily ba­sis could be valu­able for catch­ing cases and iso­lat­ing them be­fore they be­come a larger out­break. Highthroug­h­put se­quenc­ing would dra­mat­i­cally boost the num­ber of as­says — ac­tual di­ag­nos­tic tests — On­tario could process on a given day.

But “as we break one bot­tle­neck, we’re try­ing to cre­ate so­lu­tions that solve the next bot­tle­neck,” Wrana says.

“This solves the kind of end­point bot­tle­neck in the pro­duc­tion line. And now it cre­ates a new bot­tle­neck — how do you get that many pa­tient sam­ples from peo­ple? And how do you process them in the lab?”

Swabs alone are a big prob­lem. Labs have val­i­dated that test­ing us­ing a spe­cific type of na­sopha­ryn­geal swab pro­duces highly ac­cu­rate re­sults. But that leaves us re­liant on hav­ing a steady sup­ply of them — On­tario has had trou­ble ob­tain­ing a re­li­able swab sup­ply.

It also re­quires many, many health-care work­ers trained to do the some­times-un­com­fort­able in­ser­tion pro­ce­dure. All those health-care work­ers need per­sonal pro­tec­tive equip­ment too, which has its own sup­ply­chain chal­lenges. Re­searchers have tested whether saliva would work just as well — which would be help­ful, since most peo­ple can eas­ily spit into a sam­ple col­lec­tion cup.

In Toronto, a group of doc­tors and sci­en­tists tested 91 hos­pi­tal in-pa­tients at reg­u­lar in­ter­vals us­ing both na­sopha­ryn­geal swabs and saliva. They found that saliva was nearly as sen­si­tive as swabs ear­lier on in ill­ness, but was less ac­cu­rate later on.

But even if saliva sam­ples could re­place swabs, other chal­lenges re­main. How do you get them all to the lab? Wrana notes that Ama­zon or Canada Post might have the nec­es­sary in­fra­struc­ture, but we can’t just pop a bunch of virus-in­fected spit in the mail.

A virus-in­ac­ti­vat­ing so­lu­tion would have to be added, “so now the sam­ple can be treated in a more reg­u­lar man­ner, with­out the re­quire­ment for, you know, at least level two bio­haz­ard con­tain­ment.”

The next-gen­er­a­tion plat­form could “read­ily ac­com­mo­date” 50,000 to 100,000 tests per run, Wrana notes, but the chal­lenge be­comes “feed­ing the ma­chine.”

The lo­gis­ti­cal chal­lenges out­side the lab are sig­nif­i­cant, but not in­sur­mount­able.

“No­body’s ever re­ally had this prob­lem to solve be­fore, right? I think it’s solv­able. It just re­quires a part­ner­ship, co-op­er­a­tion from mul­ti­ple stake­hold­ers … gov­ern­ment, prob­a­bly some pri­vate in­dus­try, and aca­demics.”

What next?

Wrana cau­tions that there is a lot of work ahead. He also says there are other peo­ple in the re­search com­mu­nity ex­am­in­ing “equally cre­ative, in­no­va­tive ways of de­tect­ing the virus.” Suc­cess is more likely if mul­ti­ple groups are tack­ling the same prob­lem, even if they don’t all turn out to be “win­ners.”

Hav­ing a ro­bust sci­en­tific com­mu­nity in Canada is also a pre­req­ui­site.

“Di­ver­sity of re­search pro­vides a kind of reser­voir that you can draw on, in crises like this.”

COVID: Front-line thinkers is part of a reg­u­lar se­ries high­light­ing COVID-19 re­search in Canada

Jeff Wrana is a se­nior in­ves­ti­ga­tor at Si­nai Health’s Lunen­feldTa­nen­baum Re­search In­sti­tute.


When the pan­demic hit, sci­en­tists like Jeff Wrana be­gan look­ing at ways they could help. Wrana be­lieves ul­tra-high-through­put next-gen­er­a­tion se­quenc­ing, which maps DNA in­cred­i­bly quickly, can be re­pur­posed for COVID-19 test­ing.

Newspapers in English

Newspapers from Canada

© PressReader. All rights reserved.