Robert Boorstein on challenges of local NGS

 

https://www.linkedin.com/posts/robert-boorstein-aa74931_illuminas-push-to-democratize-genomics-with-activity-7311839487347572736-z2c0/ 

ROBERT BOORSTEIN

At the end of 2024, low throughput instruments (MiSeq, MiniSeq, iSeq) represented 52% of placements, but only 10% of the pull-through revenue. This compares with 10% of placements, 70% of revenue for high throughput instruments (NovaSeq), and 26% of placements and 21% of sales for mid throughput instruments (NextSeq). (Data abstracted from Nephron Research)

I suspect that Illumina has been reluctant to pull the plug on its lower throughput instruments, in part because they most likely provide significantly higher margins. (Large customers have more negotiating leverage than lower throughput ones).  No company wants to abandon 50% of its users. Unfortunately, the combined output of those users at least in clinical diagnostics, is a minimal portion of overall clinical demand, and it can be readily absorbed by laboratories with high throughput platforms.
Illumina has long advocated for the democratization of NGS. https://lnkd.in/eCvBgctg. [Gen Eng News]

They now seem to be abandoning that strategy. At least for clinical diagnostics, the rationale for small labs running NGS has made minimal economic sense.  For small throughput labs--the reagent cost per test, the labor cost per test, the validation cost per test, he professional costs per test, and the capital cost per test—all are much higher (up to 10X higher ) than for labs running high volumes on high throughput instruments.

For NGS to make sense to be democratized into smaller clinical labs, there needs to be a clinical need (time sensitivity, local knowledge) to run specimens locally rather than send them to large central laboratories. In addition, the instruments cannot just be smaller versions of large instruments, but they need to be designed with ready-to-run content, and minimal validation,  minimal labor, and professional costs. Costs cannot increase prohibitively when samples are run in small numbers to provide rapid turnaround time.
Companies such as Cepheid or Biofire have produced instruments in wide use in diverse clinical settings to provide sophisticated molecular diagnostics precisely because the costs of operation have been minimized compared with the ongoing reagent cost.   Companies wishing to democratize NGS testing need to identify both clinical needs that benefit from local testing, and design systems that minimize the cost of operations, validation, and regulatory compliance.

###

BRUCE QUINN

Callback to the 2018 News article. [GenEng News 2018]

How does the FDA Trusight Comprehensive (for CGP in tumors) fit the issues raised?

####

TOUMY G.  CSO

“Democratizing sequencing” was actually first pushed by IonTorrent, then later by Illumina and ONT. This was always a fallacy, because in addition to the reasons you mentioned in your post, none of these companies provided easy, out of the box solutions for secondary or tertiary analysis. Hence a lot of the instruments ended up as doorstops in both research and clinical labs.

ANDY WATSON

My bet is on the production scale labs for clinical genomics - doesn’t take long to ship samples, little delay to wait for a full ‘batch’, domain expertise, quality systems, data handling & analytics, and of course the cost. Localized testing seems to be more driven by a desire for revenue capture by the hospital systems and other rules ex-US - not really rational big picture drivers.

ROBERT BOORSTEIN

Thank you, Bruce, for your question, and Andy and Toumy for your frank and relevant comments.  The Illumina Trusight FDA CGS is still a complex and cumbersome process, requiring 4-5 days after extraction, and 10 hours of analysis time. (Data from Illumina marketing)

 The problem is that there is a steep cost vs turnaround time tradeoff based on batch size, both for reagents and for labor.  Assume a lab runs 250 tests a year, or 5 per week, and the batch size is 10. If you run full batches, you add a full week to the average turnaround time, more than the analysis itself. I estimate that the average user of Illumina MiSeq/MiniSeq is doing less than 250 tests a year, extrapolating from published data (Nephron Research). Large labs running larger instruments will still have major cost/test advantages in labor, reagents and validation.

 As Toumy points out, many small labs have unused NGS systems, and Illumina is now sunsetting the majority of these product lines. My observation is that many labs did not have realistic business plans, and severely underestimated the time and cost of validations, the cost of recruiting and maintaining skilled staff, and the difficulty of obtaining paying business.