I was given a questionairre on the future of clinical sequencing in the next decade - whether FFPE, or MRD LBx, and so on. Of course, a projections ten years ahead are tricky. What might we have predicted in 2000, or 2010, and how would reality have looked by comparison?
Three clinical landmarks might be introduction of the Oncotype Dx test around 2005, the widespread acceptance of BRCA sequencing (rather than hot spots by circa 2010, and the clinical launch of Foundation Medicine around 2012.
But it would be hard to imagine that first decade of snowballing clinical sequencing without also considering the billions in fund-raising and investments (both R&D and manufacturing) by Illumina (and others). There wasn't some secular trend to10X as much clinical sequencing without all those investments.
Similarly, today, we will want to project growth of MRD testing, etc, but it's hard to see it as a secular trend somehow independent of investments by Natera, Guardant, Freenome, Caris, and many others. (Add in Pac Bio, Thermo Fisher, Roche, 10X, Owkin, and others). If industry funding cut back substantially, the pace of clinical science & growth would slow, and vice versa.
I asked Chat GPT what it makes of these ideas, results below.
###
 |
| ILMN Market Cap |
###
Chat GPT:
Illumina’s 2000 IPO filing reads today like a document from a different scientific era—one in which the central problem in biology was not yet simply how to scale sequencing into the clinic, but more broadly how to industrialize the measurement of biology at all. In its S-1, Illumina did not present itself primarily as a sequencing company in the modern sense. Instead, it described a platform business built around BeadArray and Oligator technologies, targeting SNP genotyping, gene expression profiling, proteomics, and high-throughput screening. The underlying thesis was that biology was entering a phase of massive data generation, and that the winning company would be the one that solved the bottleneck of throughput and cost across multiple assay types. That framing matters. Illumina was not merely forecasting demand; it was proposing to help build the infrastructure that would make such demand possible.
The growth expectations in the IPO filing were correspondingly ambitious. Illumina argued that the rapid accumulation of genomic information—then driven in part by the recent Human Genome Project—would catalyze new markets in therapeutics and diagnostics. It suggested that SNP analysis alone could eventually require billions of assays annually, and that gene expression technologies might one day move into routine clinical use as costs declined and genomic knowledge expanded. The company projected near-term product launches in genotyping and expression analysis and envisioned a pipeline of increasingly powerful systems. Implicit in these claims was a belief that genomics would move from a specialized research activity toward an industrial process, and that Illumina could become one of the key suppliers enabling that transition.
At the same time, the risks outlined in the S-1 were stark and familiar for an early-stage life science tools company. Illumina had limited commercial revenue, having derived much of its early support from government grants, and it had accumulated operating losses. It depended on key partners, including PE Biosystems, for important pieces of instrumentation, reagents, and commercialization. It warned investors that it had only limited high-volume manufacturing experience, that some critical components came from single-source suppliers, and that competitors might outpace it technologically. The competitive set included firms such as Affymetrix and Agilent, which were already well established in adjacent areas of arrays and instrumentation. In other words, the filing makes clear that Illumina was raising capital not to exploit a settled market position, but to survive and scale in a technological race whose outcome was still very uncertain.
With hindsight, one can see that the most consequential element not yet central to the 2000 narrative was next-generation sequencing itself. That pivot came later, especially with Illumina’s 2007 acquisition of Solexa, which provided the sequencing-by-synthesis platform that would come to dominate short-read genomics. This later history highlights an important analytical point: it is misleading to treat the growth of genomics from 2000 to 2010 or 2020 as an external trend that companies merely rode.
Illumina was not just a beneficiary of the genomics boom. It was one of the companies that materially accelerated it. By investing heavily in chemistry, instrumentation, consumables manufacturing, software, and global support infrastructure, Illumina helped reduce the cost and increase the reliability of sequencing to the point that large-scale research and clinical adoption became feasible. In that sense, the market did not simply grow around Illumina; Illumina helped create the conditions under which the market could grow.
 |
| ILMN annual R&D (GAAP financials) Total about $10B, last two years ~$2B. |
This becomes even clearer when contrasted with the earlier dominance of Applied Biosystems, later part of Life Technologies and now within Thermo Fisher Scientific. Applied Biosystems’ capillary electrophoresis instruments, especially the ABI 3700, were among the workhorses of the Human Genome Project and the competing private effort led by Celera. Those systems industrialized Sanger sequencing and made the first human genome possible. In that respect, ABI played a historic and indispensable role. However, its strength belonged largely to a sequencing paradigm that scaled more slowly and at much higher cost. Illumina’s rise was tied to a different model: massively parallel sequencing, with dramatically increasing throughput and sharply falling cost per base.
It would be too simple to say that Applied Biosystems was merely “left behind.” The company adapted and continued to innovate, including through next-generation platforms such as SOLiD, and it remained important in sequencing and related markets. But in the core domain of high-throughput short-read sequencing, Illumina achieved a level of dominance that effectively redefined the field. The transition from ABI’s capillary systems to Illumina’s sequencing-by-synthesis platforms marks not just a shift in market share, but a shift in technological regime.
The broader lesson is that markets of this kind cannot be understood by straight-line extrapolation alone. The trajectory of clinical and research genomics over the last two decades was not preordained in 2000. It depended in large part on firms willing to make large, uncertain investments in platform development, manufacturing scale, and commercial infrastructure. Illumina’s IPO filing shows a company asking investors to fund precisely that kind of risk-taking. In retrospect, those investments did more than reward shareholders. They helped accelerate the timetable on which genomics became a routine part of both scientific research and clinical medicine.
 |
| click to enlarge |
##
Thanks for reviewing what was expected, and what really happened, circa 2005-2025. How about today? What does this inform us as we think about 2027-2037?
###
The Next Double Helix in Precision Medicine: Clinical Adoption and Strategic Investment, 2027–2037
If the core genomics thesis of 2000–2020 was that biology would become clinically useful once sequencing became cheap enough, scalable enough, and operationally reliable enough, the thesis for 2027–2037 is more complex. The question is no longer whether genomics matters. It already does. Comprehensive genomic profiling, or CGP, now has a real foothold in oncology practice, especially in advanced and metastatic cancer, where it is increasingly embedded in diagnostic workflows, biomarker selection, and molecular tumor board decision-making. In the United States, recent literature describes CGP as standard clinical practice for many advanced cancers, while ESMO has also continued to formalize molecular tumor board frameworks that assume broad genomic data will increasingly be part of real-world cancer management. (1-3) (Nature)
But the future decade will not be driven by a single secular trend line. It will be driven by the interaction of two intertwined forces—a kind of strategic and clinical double helix. One strand is the familiar secular progression of medicine itself: more biomarker-driven care, more inclusion of molecular profiling in guidelines, more routine use of liquid biopsy, and more standardization by payers, pathology groups, and oncology networks. The other strand is the investment-dependent buildout of new platforms: computational pathology, AI-enabled image analysis, spatial biology, multimodal biomarkers, and AI-guided genomics. These fields will not mature simply because clinicians want them. They will mature only if companies continue to fund instrumentation, software, assay development, validation studies, regulatory work, commercial infrastructure, and manufacturing scale. The key lesson from Illumina’s earlier history still applies: markets like this do not merely “happen.” They are partly built into existence by companies willing to invest ahead of the demand curve. (Nature)
 |
| Natera R&D investments, 2015-2025 Last 5 years about $2B |
That is why the comparison to the earlier genomics era is so instructive. Looking backward, it would be a mistake to describe the clinical and commercial genomics boom of 2010–2020 as though Illumina simply surfed a rising wave. Illumina helped create the wave by pouring capital into chemistry, instrumentation, consumables, software, installed-base support, and manufacturing capacity. The future may look similar, except that the enabling technologies are no longer only sequencing. The next decade may instead revolve around whether investors and operators can industrialize a broader multimodal measurement stack: sequence, image, tissue architecture, spatial localization, and AI-based interpretation, all linked to clinical action. That is not just a scientific challenge. It is also a capital allocation challenge. (Nature)
Within that future stack, CGP looks like the most mature pillar. It is already clinically useful and increasingly routine in advanced oncology. The more interesting questions for 2027–2037 are about the adjacent growth curves. MRD appears to be the most plausible next expansion. Medicare’s MolDX program already has coverage infrastructure for MRD testing in cancer, which means the field has moved beyond pure aspiration and into the phase of selective, operational reimbursement. That does not mean MRD is fully normalized. Coverage remains conditional, evidence standards are still evolving, and many use cases remain under active debate. But this is exactly what an early platform market looks like: enough coverage and enough usage to support further investment, further evidence generation, and further clinical entrenchment. (4-5) (Centers for Medicare & Medicaid Services)
MCED, by contrast, remains the most tantalizing but unstable frontier. On the positive side, the policy environment improved in early 2026 when federal legislation created a clearer Medicare coverage pathway for FDA-approved MCED tests. That matters because it gives industry and investors a more legible reimbursement horizon. But the clinical and commercial picture is still unsettled. The NHS-Galleri study failed to meet its primary endpoint for stage shift, which immediately reminded the field that plausible biology and exciting commercialization do not automatically translate into population-level clinical utility. So MCED remains important, perhaps enormously important, but not yet inevitable in the way CGP increasingly appears to be. It is still both evidence-sensitive, and capital-sensitive, as a future market. (6-7) (Exact Sciences)
The same “double helix” logic applies to computational pathology and AI-enabled diagnostics. These fields are real, but still in the translational stage. The FDA’s AI-enabled device list now runs into the thousands across all categories, showing that AI in medical devices is no longer exotic. Yet digital pathology remains a much earlier subfield; one 2025 review noted that, as of August 2025, only three AI/ML software devices in digital pathology had been cleared. That gap is revealing. It suggests not that digital pathology lacks promise, but that the pathway from technical performance to scalable clinical platform is demanding. Success will require not just algorithms, but image standardization, scanner deployment, workflow integration, validation, reimbursement logic, and trust from pathologists and health systems. (8-9) (U.S. Food and Drug Administration)
Spatial genomics and AI biology may represent an even earlier version of the same story. 10x Genomics is explicitly trying to connect single-cell and spatial profiling to clinically relevant tumor maps and diagnostic applications, while Owkin is building multimodal AI systems intended to combine pathology, genomics, and spatial data into biomarker and decision engines. Whether these efforts become mainstream clinical infrastructure by the early 2030s is still uncertain. But the strategic shape of the problem is already visible: without large and sustained investment, these technologies may remain research tools; with sustained investment and successful validation, the future itself changes, and they become part of the next clinical platform layer. (10-12) (BioSpace)
Sidebar: Same Themes in Robotic Surgery
This duality is not unique to genomics. Robotic surgery offers a parallel case study. It would be misleading to describe the growth of robotic surgery as a simple function of surgeon demand or demographic need. The rise of the da Vinci ecosystem has depended on continuous product improvement, training programs, commercial deployment, installed-base expansion, manufacturing scale, and years of capital-intensive reinvestment by Intuitive. As of the end of 2025, Intuitive reported more than 11,000 da Vinci systems installed globally and more than 3.1 million da Vinci procedures in 2025 alone. That is not just a secular trend. It is the result of a company repeatedly funding the platform that enables the trend. The same may prove true in precision medicine over the next decade. (13-14) (Intuitive Surgical)
Similarly, Apple didn't just make more glass phones in response to its observations about the market; the huge investment in R&D and manufacturing made the falling prices, rising quality, and market growth possible.
So the right futurist thesis for 2027–2037 is not merely that precision medicine will keep growing. It is that growth will be strongest where clinical adoption and strategic investment reinforce one another. CGP is already the mature backbone. MRD looks like the most credible adjacent expansion. MCED remains a high-upside but still unstable bet. Computational pathology, spatial biology, and AI genomics may become the next major platforms—but only if capital, validation, reimbursement, and workflow all arrive together. In that sense, the future will not be shaped by secular trends alone. It will be shaped by the continuing dance between medicine’s appetite for better tools and industry’s willingness to build them. (Nature)
References
(1) Carr A, et al. “Tumor diagnosis recharacterization enabled by comprehensive genomic profiling in real-world clinical practice.” npj Precision Oncology (2025). Nature. (Nature)
(2) Huang CY, et al. “Integrating comprehensive genomic profiling in the management of oncology patients.” Current Oncology (2025). PubMed Central. (PMC)
(3) European Society for Medical Oncology. “Treatment Guided by Comprehensive Genomic Profiling and Expert MTB Decision-Making Improves Outcomes in Pretreated Patients with Metastatic Cancer.” ESMO Oncology News, October 14, 2025. (ESMO)
(4) CMS Medicare Coverage Database. “MolDX: Minimal Residual Disease Testing for Cancer (LCD L38814).” Centers for Medicare & Medicaid Services. (Centers for Medicare & Medicaid Services)
(5) CMS Medicare Coverage Database. “Billing and Coding: MolDX: Minimal Residual Disease Testing for Solid Tumor Cancers (Article A58434).” Centers for Medicare & Medicaid Services, revised 2025–2026. (Centers for Medicare & Medicaid Services)
(6) Exact Sciences. “Exact Sciences Applauds Passage of Legislation Establishing Medicare Coverage Pathway for Multi-Cancer Early Detection Tests.” Press release, February 3, 2026. (Exact Sciences)
(7) Ernst SW. “NHS-Galleri MCD trial fails to meet primary endpoint of stage shift.” The Cancer Letter, March 6, 2026. Additional insurance-context summary from RGA, March 24, 2026. (The Cancer Letter)
(8) U.S. Food and Drug Administration. “Artificial Intelligence-Enabled Medical Devices.” FDA, updated 2026. (U.S. Food and Drug Administration)
(9) Makhlouf HR, et al. “Digital pathology imaging artificial intelligence in cancer diagnosis and treatment.” Review article, 2025. PubMed Central. (PMC)
(10) 10x Genomics. “10x Genomics to Advance Research in Diagnostic Applications of Tumor Single Cell and Spatial Profiling.” Press release, January 13, 2026. (BioSpace)
(11) Owkin. “AI Biomarkers and Precision Medicine Advancements.” Company overview page. (Owkin)
(12) Owkin. “Owkin Brings Biology Super Intelligence Closer to Reality with New AI Infrastructure for Biological Breakthroughs.” Press release, January 12, 2026. See also Owkin spatial-biology conference materials. (Owkin)
(13) Intuitive Surgical. “Intuitive Announces Fourth Quarter Earnings.” January 22, 2026. Installed base of 11,106 da Vinci systems as of December 31, 2025. (Intuitive Surgical)
(14) Intuitive Surgical. “Intuitive Announces Preliminary Fourth Quarter and Full Year 2025 Results.” January 14, 2026. Approximately 3,153,000 da Vinci procedures in 2025. (Intuitive Surgical)
