NEW YORK – NanoString Technologies is betting even bigger on spatial genomics and proteomics, revealing on Tuesday a new platform that it claims can provide subcellular resolution for RNA and proteins in tissue samples.
In a presentation for investors and Wall Street analysts, NanoString officials touted the new spatial molecular imager (SMI), an instrument that uses the firm's proprietary Hyb & Seq chemistry for in situ analysis of at least 1,000 RNAs and, theoretically, the same number of proteins. It's a tool that will complement its GeoMx Digital Spatial Profiler, offering high-resolution and targeted analysis of genes and proteins of interest in discovery, translational, and possibly even clinical applications. The new platform will work with both formalin-fixed paraffin-embedded and fresh-frozen tissue slides and can analyze molecules across the entire slide down to localized areas within cells, including the nucleus.
With GeoMx and SMI, NanoString is pushing hard to offer solutions for researchers in the spatial analysis market, which is even bigger than previously determined, NanoString CEO Brad Gray said. Based on new market research, he estimated the total addressable market for spatial biology is at least $12 billion. With SMI, NanoString hopes to meet the needs of segments of the market not currently met by GeoMx, including diagnostic testing potentially worth $6 billion per year.
Copying the product launch strategy developed with GeoMx, NanoString is planning to make SMI available to select customers in early 2021 through its technology access program (TAP), distribute a small number of instruments in early 2022, and fully launch the instruments by the second half of 2022. NanoString officials noted that "spatial molecular imager" is a description of the instrument and not the brand name for the new platform.
"Spatial molecular imaging opens up new possibilities to understand the diversity of T cells infiltrating a tumor at the single-cell level without dissociating tissue," Evan Newell, a T-cell researcher at the Fred Hutchinson Cancer Research Center, said in a statement. His lab has already begun collecting data from SMI through NanoString's TAP, looking at the location of different T-cell subsets and their interaction with tumors. "We're very excited about the capacity to look at so much detail in tissue," he said in a video segment shown during the investor event. "This will give us a huge opportunity to see exactly where the different T-cell subsets are — which ones are interacting with the tumor, which ones are excluded from the tumor. From this, we can get a lot of insight as to which are the most important for controlling tumors."
"We have some beautiful data that's showing the sub-diffraction limit resolution mRNA expression levels of all the cells in [FFPE] tissue," he said.
Investors responded favorably to the news; in Tuesday trading on the Nasdaq, shares of NanoString rose 10 percent to $54.63. "We walked away from the meeting incrementally positive about NanoString's positioning in spatial biology … and the growth runway," JP Morgan analyst Tycho Peterson wrote in a research note.
According to NanoString officials, the Seattle-based firm turned its Hyb & Seq chemistry towards in situ analysis in June 2019. Previously, the firm had planned to use it as part of a clinical sequencing platform. Launched in 2016 at that year's Advances in Genome Biology and Technology conference, Hyb & Seq represented a play into targeted sequencing built on NanoString's optical barcoding technology already at work in its nCounter line of products.
As originally conceived, optically barcoded Hyb & Seq probes hybridized to six base-long regions of captured DNA and RNA targets. The instrument would then read out the barcodes of probes interacting with their targets in cycles to identify bases and their positions. Hyb & Seq probes were washed away and the cycle was repeated until the captured targets were sequenced. The chemistry did not use enzymes or amplification.
But as the clinical sequencing market coalesced around Illumina's platforms and NanoString delved into spatial biology with GeoMx, Gray looked for a new way to deploy Hyb & Seq. "[SMI] has really become the focus of our Hyb & Seq program," Gray said. "Clinical sequencing is not a priority," although the firm is still in discussions with potential partners to develop such a platform, he said.
In the SMI, Hyb & Seq probes do not target unknown, captured nucleic acids, rather, they target single-stranded barcode tags appended to in situ hybridization RNA probes or antibody-labeled DNA-nucleic acid probes. "With SMI, we put the Hyb & Seq chemistry into the tissue directly, to ID the RNA or protein barcode directly inside the tissue with a spatial resolution determined by the tissue itself. By doing so we can achieve much higher spatial resolution inside the tissues," Gray said. Sample prep is similar to many other in situ hybridization protocols. "You take that slide and you put it into the spatial molecular imager, close the door and you're done," NanoString CSO and Senior VP of Research and Development Joe Beechem said. "Everything else is automated." SMI will detect proteins using antibody reagents developed internally, or in partnership with AbCam. "Any RNA or protein content developed for GeoMx can be converted into an SMI in-situ imaging reagent," he said.
NanoString will begin by offering a proprietary 1,013 gene panel that helps single-cell researchers address specific questions about cells, including their type, activation state, and signaling with neighboring cells. The platform is "capable of unlimited multiplexing for both RNA and proteins," Beechem said. "The protein limitation is only related to the number of validated antibodies," which currently number more than 6,000, he said, adding that protein panels NanoString will offer likely will be lower-plex than for RNA, mostly due to demand and cost.
"I believe there will be other 1,000-plexes we will make designed for different tissue types," Beechem said.
In experiments to test SMI's sensitivity, NanoString researchers compared the platform to single-cell RNA expression data from studies conducted at the Broad Institute for the Cancer Cell Line Encyclopedia. Analysis of 2,000 cells using SMI yielded 83 percent of the RNAs detected by RNA-seq of about 500,000 single cells, Beechem said.
The optics inside the box are proprietary and custom built to the new platform, using four-color fluorescent imaging on each cycle. Beechem declined to provide exact figures on resolution at this time, although it is presumably less than 10 micrometers, the resolution limit of GeoMx, and potentially below 1 micrometer, as suggested by presentation slides.
Data are presented as molecules with coordinates in 3D. A software suite groups molecules into cells and those cells are arranged in spatial context with each other.
Gray said the firm would provide more technical details at the upcoming AGBT virtual meeting in March 2021.
Wide-open market
NanoString joins several other competitors in what Gray characterized as a wide-open spatial analysis market. While several other single-cell resolution in situ analysis technologies exist, none appear ready for prime time. So far this year, Vizgen, a Harvard University spinout that is commercializing multiplex error-robust fluorescence in situ hybridization (MERFISH), has emerged from stealth, and 10x Genomics has announced it will buy ReadCoor, another Harvard spinout offering in situ multiomics analysis, as well as Cartana, an in situ gene expression analysis company. In October, 10x CEO Serge Saxonov told GenomeWeb that the firm would be investing more into the ReadCoor platform and declined to provide a development timeline.
Whoever's ready soonest will have first-mover advantage in a market that NanoString says is worth even more than it initially thought. Gray said the firm has worked with a management consulting firm to do market research on the spatial biology market, concluding that it is worth at least $12 billion per year, more than six times its previous estimate from early 2019.
Previously, NanoString had estimated that discovery and translational research markets would each be worth about $1 billion. Now, with expanded capabilities and increasing momentum, each of those research markets could be worth $3 billion.
Half of that is met by wide-angle technologies like GeoMx, he said, while SMI will address the remaining half. NanoString officials stressed that GeoMx and SMI are not substitutes for each other.
"Both are standalone systems, but they generate unique sets of information," Beechem said. "When you have both, you can span the entire dynamic range to answer any spatial biology question." Moreover, data sets generated by the different platforms will harmonize in downstream analysis.
Included in the increased TAM is the opportunity to create new spatial omics-based molecular diagnostics, potentially worth $6 billion, or more. "To size this market, we looked at diagnostic tests that might benefit from a spatial component," he said. Next-generation sequencing, tissue-based diagnostics, and flow cytometry-based diagnostic methods combined to represent about $23 billion in current annual spending. "If 25 percent of those tests move over to a spatial format, we'll have a $6 billion spatial diagnostic TAM," Gray said. "And this may be a conservative estimate. Note, we haven't included other areas of molecular diagnostics or novel spatial technologies and tests that may be discovered by our customers."
Gray also noted that spatial gene expression information is simply more valuable than gene expression information alone, as measured in consumables costs. A 770 target nCounter assay provides about $250 in value to NanoString per sample, he said, while a much smaller 84 target GeoMx assay with nCounter readout is worth about twice that at $500 per sample. GeoMx with NGS readout is even more valuable; the Cancer Transcriptome Atlas provides $1,250 per sample of value to the firm.
To get SMI into customer labs, NanoString will follow the commercial roadmap it established with GeoMx. The firm will begin offering its new spatial imaging platform as a service beginning in the first half of 2021. In early 2022 it will launch beta access to instruments followed by a full commercial launch in the second half of 2022. "It's the perfect game plan to follow," NanoString Senior VP of Sales and Marketing Chad Brown said.
Gray said it was too early to discuss instrument and consumables pricing, however, he suggested that NanoString would follow the GeoMx plan here, too, where the instrument price point would not pose a major barrier to adoption.
And NanoString plans to have flexible pricing to entice the right customers to join its TAP, who could help show what's possible with the system.