23andMe Shifts Strategy Away from NGS to Focus on Scalable Consumer Genomics and Drug Discovery

23andMeThe abandonment of next-generation sequencing (NGS) R&D came as somewhat of a surprise, effectively dousing cold water on the once-hot gene testing technology. However, upon closer reflection, it is not entirely unexpected. The immaturity of the technology, high sequencing costs, and low consumer willingness and demand for sequencing have already led to skepticism about the future of NGS. This is also evident from the reaction of foreign media to this event, which did not cause the same uproar as when the FDA required 23andMe to suspend providing health-related genetic testing services to new users in 2013.

As verified by international media outlets such as TechCrunch,23andMeThe laboratory has halted its NGS project, and at least five lab technicians have been laid off, including Dr. Jill Hagenkord, the Chief Medical Officer who had overseen the project for two years. Founder Anne Wojcicki publicly stated that this decision was unrelated to slowing sequencing demand, funding issues, or regulatory concerns. “Our funding is ample; it’s just that the situation has become somewhat complex.” So, why23andMeIs NGS Being Abandoned? What’s Really Happening in the Genetic Testing Market? VCBeat (WeChat: vcbeat) provides an analysis and explores its future with you.

In the field of genetic testing, 23andMe has long enjoyed a prestigious reputation. From its founder, Anne Wojcicki, to its investors, including Google, Johnson & Johnson, and the NIH, the company has been surrounded by a star-studded aura. In 2008, it was honored with Time magazine’s “Best Invention of the Year” award, basking in immense glory at the time. From securing $8.95 million in Series A financing from companies such as Google in October 2007, to receiving $1.7 million in funding from the NIH this October, 23andMe has completed seven rounds of financing to date, raising a total of over $240 million.

Given such strong favor from the capital market, what has been 23andMe’s core business all along? It is the SNP-based (single-nucleotide polymorphism, referring to DNA sequence variations caused by single nucleotide changes at the genomic level) detection of 650,000 loci in the human genome, facilitating large-scale, voluntary population genetics sampling and genotyping. In the field of genetic testing, this constitutes a relatively basic service.

23andMe does not employ whole-genome sequencing; instead, it utilizes the Illumina-based HumanOmniExpress-24 genotyping array for screening. This approach not only effectively reduces equipment and R&D costs but also lowers medical risks. The test achieves an accuracy rate of up to 99.7%, while being more cost-effective than next-generation sequencing. SNP genotyping offers high accuracy, strong flexibility, high throughput, short turnaround time, and low cost, along with advantages in data analysis that are rapid, accurate, and standardized.

What is the price of sequencing? In 2003, the cost of the first human genome test was as high as $2.7 billion. 23andMe’s initial pricing was $999, which has steadily declined over time. After securing more than $60 million in Series B and C funding rounds, the test price dropped to $299. In 2013, the FDA halted 23andMe’s service of providing health guidance to users based on genetic information due to the lack of established regulatory standards, causing the company to slash its testing fee to a rock-bottom price of just $99. Currently, the typical price is only $149 or $199, offering a clear advantage compared to the consistently high cost of next-generation sequencing (NGS), which has remained around $1,000 for the past three years.

To date, 23andMe has likely accumulated a user base of more than 1.5 million, establishing what can be described as the world’s largest database of gene-phenotype associations. At this juncture, 23andMe faces two strategic paths: one is to continuously advance new genetic sequencing technologies and pioneer a new era of innovation; the other is to persist in selling its basic genetic sequencing services directly to individual consumers. Clearly, 23andMe has chosen the latter.

Behind such a massive genetic database lies immense financial potential. Currently, 23andMe’s new vision is to transform the traditional drug development model by “engaging patients in drug development.” Leveraging big genetic data, large pharmaceutical companies develop treatment plans for patients, while patients provide the data required for research.

In March last year, 23andMe established a medical team led by Richard Scheller, the former CEO of Genentech. Scheller serves as the Chief Scientist of 23andMe’s medical team and is responsible for leveraging the genetic data already collected from 850,000 customers to advance the development of new therapies for certain rare diseases. This marks 23andMe’s formal entry into the pharmaceutical industry. Genentech spent a total of $60 million to purchase data on 3,000 Parkinson’s disease patients from 23andMe, averaging $20,000 per customer contributed to 23andMe.

In its collaboration with Pfizer, 23andMe grants Pfizer access to its research platform, including 23andMe’s services and genetic data analysis from a population of over 800,000 individuals. Within this extensive database, more than 80% of testers (approximately 650,000 people) have consented to participate in research. During the initial phase of the partnership, Pfizer will analyze data from 5,000 lupus patients sourced from 23andMe to further investigate the genetics of lupus.

Genentech paid $10 million to acquire DNA sequences from Parkinson’s disease patients. The two parties will jointly analyze genomic sequencing data from 3,000 Parkinson’s disease patients, aiming to identify new therapeutic approaches for this neurodegenerative disorder. Under this collaboration, 23andMe will be responsible for collecting data from Parkinson’s disease patients and conducting genomic sequencing, while Genentech will leverage this information to develop potential treatment strategies and drugs.

In addition, the FDA has approved another single-health product from 23andMe for predicting Bloom syndrome. To date, publicly available online data indicate that 23andMe has entered into collaborations of varying scales with 13 pharmaceutical companies. Anne Wojcicki, CEO of 23andMe, stated that she hopes everyone will in the future be able to access 23andMe’s vast genetic database to identify common factors among groups of individuals with the same disease through gene sequencing, thereby determining therapeutic targets and even predicting the likelihood of disease onset in specific populations.

Commenting on the matter, Barbara Evans, a legal expert at the University of Houston, stated that monetizing DNA data is an excellent revenue-generating strategy. However, drug development is inherently a high-risk field with an exceptionally lengthy process. On average, developing a new drug costs approximately $2 billion and takes more than 10 years to bring to market. Moreover, 23andMe’s services are currently limited to providing information on only a dozen rare diseases. Although 23andMe’s entry into the pharmaceutical sector has addressed its profitability challenges, its future prospects remain to be seen.

In addition to the high costs, NGS (Next Generation Sequencing), also known as "next-generation sequencing," faces challenges related to technological maturity, which is likely one of the reasons why 23andMe abandoned it. Wojcicki stated that she was uncertain whether next-generation sequencing, a currently more complex and expensive test, could attract a significant number of customers and gain consumer acceptance. "Most people are not even aware that these types of tests exist," she said. "I believe the market is still in its early stages." So, what is the current state of development for NGS?

Since the launch of the GS 20, the world’s first commercial second-generation DNA sequencer, in 2005, the gene industry has embarked on its development journey. Currently, platforms from Illumina and Thermo Fisher Scientific account for approximately 90% of the global market share. From a technical perspective, next-generation sequencing (NGS) can be applied across various fields of multi-omics research. For instance, in genomics, whole-genome sequencing and whole-exome sequencing are used to detect somatic mutations; in epigenetics, bisulfite sequencing is employed to identify DNA methylation sites, chromatin immunoprecipitation sequencing (ChIP-Seq) is used to analyze histone modifications, while DNase-Seq and FAIRE-Seq are utilized to assess chromatin structure, and ChIA-PET is applied to investigate chromatin interactions regulated by specific transcription factors.

However, the more mainstream clinical applications of NGS are concentrated in reproductive health and oncology diagnosis and treatment. Based on the diverse needs of reproductive health, genetic sequencing can be further subdivided into preimplantation genetic testing, prenatal testing (For exampleNon-invasive prenatal testing (NIPT) and newborn disease screening. In 2016, researchers at the Children’s Hospital of Eastern Ontario in Australia conducted experiments to evaluate the feasibility of applying sequencing-based diagnostic technologies in the clinical diagnosis of genetic disorders in newborns. This study demonstrated the immense potential of next-generation sequencing (NGS) technology for efficient and rapid molecular diagnostics. NGS has indeed proven capable of accurately detecting various chromosomal variants that are not identifiable through conventional karyotyping or chromosomal microarray analysis. Furthermore, early diagnosis and treatment of cancer are critically important, and NGS gene sequencing serves as an effective approach to address this need. However, it is currently essential to standardize the operational protocols and quality standards of sequencing companies, as only standardized techniques can ensure subsequent precision diagnosis and precision therapy.

Undoubtedly, next-generation sequencing (NGS) high-throughput sequencing technology has advanced research in oncology and immunology, as well as the development of personalized immunotherapy, enhancing our understanding of cancer genomics and the intracellular mechanisms involved in tumorigenesis. However, the adoption of NGS has also introduced a series of challenges. While NGS generates massive volumes of data, its quality remains suboptimal (reportedly, the error rate during sequence assembly ranges from 0.1% to 15%), and the relatively short read lengths typical of NGS necessitate more rigorous and complex sequence assembly processes.

The maturity of gene sequencing product applications depends on the accuracy of genomic data interpretation. Currently, constrained by the short read lengths of next-generation sequencing (NGS), interpretive capabilities are primarily limited by the scarcity of genomic data based on precise disease classifications. According to a survey by Ebiotrade, 69% of respondents identified data analysis and interpretation as the most significant bottleneck hindering the development of the sequencing industry chain. The three essential elements for effective data analysis include high-performance computing platforms, specialized analytical software, and high-quality large-sample databases. Raw sequence files provided by sequencing service companies cannot yield any valid information before undergoing systematic analysis and processing. Computing platforms are used to perform a series of foundational analyses on raw sequence files generated by sequencing instruments, such as quality filtering and sequence alignment, while analytical software and large-sample databases are employed for genetic interpretation and counseling.

Therefore, as NGS aims for broad clinical application, turnaround time has become a primary challenge. For patients with severe neurological disorders or life-threatening cancers, a waiting period of several weeks for whole-genome sequencing (WGS) analysis can cause them to miss the optimal window for treatment. Furthermore, the massive volume of data, often reaching the petabyte (PB) scale, poses significant challenges for downstream processing, necessitating revolutionary solutions in data storage and bioinformatics.

It is evident that SNP-based testing primarily targets consumer-grade products, whereas NGS addresses clinical-grade applications, representing two distinct markets. When using 23andMe, users need only register an account on the official 23andMe website and order the genetic testing service to receive a saliva sample collection kit mailed by 23andMe. After mailing their saliva samples back to 23andMe and completing the payment, users can wait for a period of time to obtain an online DNA report. This report provides insights into their ancestral origins, traits and talents, cardiovascular genetic diseases, food tolerances, and drug responses.

This report presents users with their ancestral admixture composition, including proportions of European, African, and Asian ancestry, thereby pinpointing their ancestors to specific regions within particular continents. Users may also choose to submit their DNA data for genetic research purposes. Generally, individuals who consent to data submission can participate in over 230 studies, primarily focused on identifying treatments and cures for diseases.

23andMe originally provided users with data on disease and health risks. However, as this practice involved medical activities and had not received approval from the U.S. Food and Drug Administration (FDA), 23andMe received a warning letter from the FDA on November 22, 2013, prohibiting it from performing any analysis of users’ health data. Consequently, users could no longer learn about their probability of developing certain diseases through reports. Until relevant regulatory frameworks were perfected, the FDA could only ban such health reports from 23andMe.

23andMe has maintained ongoing communication with the FDA regarding the accuracy of its data interpretation, continuously securing FDA authorization through scientific and widely accepted experimental design protocols. On October 22, 2015, the FDA granted 23andMe clearance, approving in a single action 36 genetic tests for hereditary conditions, including cystic fibrosis and sickle cell anemia, as well as ancestry analysis and non-medical traits. Although the FDA has permitted 23andMe to offer direct-to-consumer genetic testing services, the currently authorized scope is limited to genetic loci associated with hereditary diseases where causal relationships are relatively well-established, excluding controversial disease risk assessments.

Anne Wojcicki hopes to continue providing disease risk analysis, but the rigorous FDA is likely to require23andMeProviding more supporting evidence requires extensive big data analysis of genuine genetic and health information. It now appears that this underlying big data has always been the core focus valued by 23andMe. Following the announcement of its withdrawal from the next-generation sequencing (NGS) business, Anne Wojcicki stated in an interview with BuzzFeed: “We have spent a lot of time on sequencing, but I believe we should gain a deeper understanding of genetic complexity. Many people still lack a basic understanding of genetic information. A major priority for the coming years is to help individuals understand their genetic foundations, so we have decided to focus on our core business.”

In terms of core business, for23andMeIn this context, obtaining rich data from a single consumer is more valuable than acquiring random genetic data from one hundred individuals. Furthermore, to expand genetic diversity, the company has turned its attention to Africa, launching the African Genetics Project this October. This initiative aims to develop new tools to help researchers explore deep genetic variation data across different population groups. 23andMe announced that it is building a reference database containing whole-genome sequences from its African American customers who have consented to participate in research. Adam Auton, Senior Scientist and Statistical Geneticist at 23andMe, stated that the company aims to include sequences from over 900 individuals in the database, which will ultimately be shared with the NIH and made available to researchers.

Discussions on “genes,” particularly cancer-related genes, have been highly active in recent days. On one hand,23andMeAs NGS is abandoned, the notion of a “precision medicine bubble” has gained renewed traction. Han Jian, a researcher at the HudsonAlpha Institute for Biotechnology in Huntsville, Alabama, and a blogger on the prominent science blog ScienceNet.cn, published an article titled “Two Key Articles That Burst the ‘Precision Medicine’ Bubble.” The first article referenced is Dr. Prasad’s “The Precision-Oncology Illusion”; the second is “Limits to Personalized Cancer Medicine” by Ian F. Tannock et al., published in the New England Journal of Medicine (2).

In particular, Dr. Prasad cited two sets of data in his article. First, a sequencing study of 2,600 patients at the MD Anderson Cancer Center found that only 6.4% of patients could be matched with appropriate targeted therapies. Similarly, in a clinical study by the National Cancer Institute, only 2% of patients harbored actionable targets for targeted drugs to date. Since not all patients with such targets respond to targeted therapies, Dr. Prasad believes that only about 1.5% of patients actually benefit from targeted treatment.

In fact, there is a crucial cognitive issue here: Is precision oncology simply synonymous with targeted therapy? Clearly not, and it is certainly not equivalent to genetic testing alone. Diagnosis should be prioritized; without precise pathological diagnosis and clinical analysis, “precision treatment” cannot exist. In summary, precision oncology has expanded beyond targeted drug therapy to encompass diagnosis, prediction of therapeutic efficacy, and monitoring of metastasis and recurrence.

Currently, there is a significant gap between domestic treatment practices in China and international standards, with the primary reason being pathological diagnosis; indeed, “diagnosis” should be prioritized above all else. However, the awkward reality is that there is a severe shortage of qualified pathologists, and tumor diagnostic reports often lack standardization and accuracy. In countries such as the United States, Japan, and those in Europe, only gene sequencing applications with clear diagnostic and clinical utility are relatively widespread, which is still far from achieving true “precision medicine.” In fact, the most pressing need at present is to increase the number of pathologists in hospitals at all levels, particularly at the grassroots level, and to enhance their fundamental diagnostic competencies and skills. Additionally, more pathology technicians should be recruited to relieve pathologists from the burdensome, trivial, and low-level task of specimen sectioning, allowing them to focus on standardizing basic routine pathological diagnostic reports and training pathology residents.

Some of the data in this article are sourced from publicly available online materials, includingTechcrunch/UNZ/Crunchbase/GENE/MarketsandMarkets / Sequencing China / BioExplorer / Singularity Network, etc.