Signify Premium Insights: Zebra Acquisition Gives Nanox Chance to Earn its Stripes

Published: November 4, 2021

This Insight is part of your subscription to Signify Premium Insights – Medical Imaging. The content is only available to companies that have subscribed to this paid-for service. To view other recent Premium Insights that are part of the service please click here.

Co-written by Steve Holloway, Bhvita Jani, Arun Gill & Dr Sanjay Parekh

Update on 20th August 2021: Since this Premium Insight was originally published, the US Food and Drug Administration (FDA) has requested additional information regarding Nanox’s regulatory submission for its Nanox.ARC multisource X-ray system.

In a filing with the US Securities and Exchange Commission (SEC) on August 12, the Israeli vendor said that a hold had been placed on the submission “pending a complete response to the FDA’s list of deficiencies”. There were no details given as to the nature of the deficiencies mentioned by the FDA, but in the filing to the SEC Nanox stated it would respond to the request within the allocated 180 days. The Israeli vendor also added that it would also continue to “optimise and develop” further features of the Nanox.ARC scanner and is even considering submitting an additional 510(k) for the next version of the multi-source Nanox.ARC in the fourth quarter of this year.

The vendor’s share price fell on the back of the news, although, as was the case for the single-source version of the Nanox.ARC, a request for additional information ultimately proved to be no barrier to FDA approval. Any delay caused by the information request will, however, impede Nanox’s ability to hit its ambitious target of deploying 15,000 Nanox.ARC scanners by the end of 2024. Delays will also blunt the vendor’s ability to reach this target with a reasonable proportion of its multi- rather than single-source scanners, a necessity if it is to deliver on its promise of “revolutionising” medical imaging.

Original Premium Insight, published 17th August 2021: Last week, another chapter was written in the erratic story of upstart modality vendor Nanox. Almost a year after the Israeli start-up listed on the NASDAQ index, the firm announced that it was making two separate acquisitions. The first, acquiring compatriot medical imaging AI developer Zebra Medical Vision, in an all-stock transaction worth $100m upfront with another $100m tied to specific milestones. The modality vendor also picked up development partner radiology service provider USARAD and its related company Medical Diagnostics Web or MDW, which operates a global network of more than 300 radiologists. Total consideration for this transaction was $30m, comprised of $21m of Nanox stock and another $9m in cash.

The Signify View

Despite its short history as a publicly-traded company, Nanox already has a colourful past. It has had short-sellers comparing the vendor to disgraced blood testing firm Theranos, awkward and opaque virtual product demonstrations at RSNA 2020, and sent mixed messages by seeking approval for what it claims is a “revolutionary product” on an FDA pathway for products ‘substantially similar’ to already approved devices. This lack of clarity has resulted in a tumultuous share price, swinging between lows of around $25, to highs of almost $90, and back again in a matter of months. No one, it seems, is quite sure exactly what the vendor is worth.

This volatility will not be stymied by the latest developments, with Nanox shares shedding as much as 17% of their value in the wake of the announcement.

Despite this reaction from the market, there are some grounds for positivity. These acquisitions will, after all, go on to create a unique proposition in the world of medical imaging: a single vendor addressing the spectrum of modality manufacture, AI software development and radiology reading services. Other vendors combine two of the three elements, with teleradiology reading groups such as Nines and Deeptek, for example combining reading services with in-house AI development, while several vendors including Siemens, Fujifilm and Canon deliver on both the modality and AI fronts. Philips has come closest to combining all three elements with its Direct Radiology acquisition, though radiology reading service assets were spun out as part of the deal, but no other vendor has combined all three in quite the same way.

This combination is intended to build what the vendor describes as “a globally connected end-to-end radiology solution and population health platform”. USARAD is a clear way to get Nanox ARC scanners into the US market. But, its in-house team of radiologists ready to read the diagnostic images is the most valuable asset today, as radiology resources become increasingly prized. Zebra’s population health tools have also found a platform partner on which to deploy, a growing concern for the business, which is struggling for traction. Combined, there is distinct potential, and a blueprint for the tightly connected unit to succeed long-term.

 Seizing the Opportunity

Realising this potential is another matter entirely. With a revenue of $5-10m last year, the acquisition of USARAD will give Nanox, which has a market cap of $1.2bn, its first revenue-generating business. As detailed in Signify Research’s report Teleradiology – World – 2021, USARAD is a sizable business, but at around 300,000 reads conducted annually, remains outside of the top 10 in the US in terms of volumes. Accounting for more than 95% of its business, the US is by far the provider’s largest market, and while covering all modalities, it targets subspecialisation, particularly MRI. The provider, which is growing, will become a US route-to-market for Nanox’s ARC scanners, although with a share of just 2% of the US teleradiology reading services market in 2020, this access won’t be transformational for the Israeli firm. In addition, Nanox will also be getting USARAD companion company, a reading service marketplace based on blockchain and reliant on cryptocurrency, which could fit well with Nanox’s medical imaging-as-a-service (MSaaS) model.

In contrast to the stability of USARAD’s steadily growing business, Zebra Medical Vision looks to be an altogether riskier gambit. As profiled in Signify Research’s Machine Learning in Medical Imaging report, the AI developer does hold some promise. It has produced seven regulator-approved products, recently attained an industry-first CTP Category III code for its vertebral compression fracture (VCF) solution and has more than 150 customers globally. However, these strengths could mask several fundamental challenges that the vendor is facing. It has, for example, not secured funding since 2018, when it secured $32.5m in a series C round, bringing its total investment to $57.4m. While a healthy amount in 2018, it does falter in comparison to some similar vendors. Aidoc, for example, secured $66m in a recent funding round alone.

 A Possessive Partner

Further, Zebra Medical is losing ground to its rivals with younger entrants such as Artrya and Cleerly making moves on cardiac health screening, for example. Consequently, the algorithm developer is transitioning away from diagnostic tools, and instead focusing on population health and population risk adjustment tools. With opportunistic screening, for example, an avenue that holds considerable potential for AI. However, to be most effective, solutions such as Zebra’s VCF detection tool need to be deployed on a large scale, their benefit is relative to the volume of diagnostic imaging procedures that they are used on. As such, even assuming it could be utilised on Nanox’s tomosynthesis systems rather than the tomography systems it was developed for and assuming Nanox meets its own targets of having 15,000 units deployed globally by the end of 2024, the vendor would still have access to a far smaller proportion of image procedures compared to a large modality and imaging IT vendor installed at many large acute hospitals and outpatient imaging centres.

For Zebra Medical, the Nanox acquisition will also likely prevent the developer partnering with other modality vendors to be deployed across broader settings. For example, with less focus on its detect and triage tools, Zebra Medical’s existing partnerships with x-ray modality vendors (e.g. Canon Medical) may become defunct. This therefore places much greater importance on the successful deployment of the Nanox Arc system for Zebra to succeed. Given the lack of evidence and increasing suspicion around the Nanox system, its 2024 target already appears speculative and over ambitious, accounting for 18.5% of the total 80,260 X-ray systems forecast to be shipped globally from 2021 to 2024. In the race for AI market adoption, it is hard to see how Zebra will be able to compete.

Worthwhile Sacrifice?

On the face of it, this acquisition will mean Zebra Medical Vision will have to make some changes to its strategy and accept some compromises, sacrificing some of its own grand ambitions to better contribute to the new broader concern of which it is a part. Zebra Medical’s failure to have a significant impact on the market so far, and the increasing competition for many of the areas addressed by Zebra Medical’s solutions, has made securing new funding difficult, leaving it vulnerable to an acquisition. The fact that it was an all-share deal with performance-related conditions also raises questions over the value of Zebra Medical, giving the AI developer a value only in relation to the volatile valuation of Nanox, rather than ascribing it a definitive cash value.

Nanox’s volatility will also raise questions about the future of the company. Some of the most scathing criticisms levelled at Nanox appear to be unfounded, with the single-source FDA approval at least demonstrating the vendor has some level of viable technology. The lack of, and often contradictory information, particularly during the vendor’s listing, has also left confusion as to which devices the Nanox ARC will compete against. Originally the Israeli’s target seemed to be computed tomography systems, with the original images showcased on its website comparable to that of CT; however, after the launch and FDA approval of its single-source Nanox.ARC it became clear that the focus was on digital radiography. Nanox’s multi-source 3D digital tomosynthesis X-ray system, filed for 510(k) clearance, will produce images more comparable to CT as a result of the multiple x-ray sources arrayed around the patient to generate 2D tomosynthesis acquisitions.

While the vendor has created a unique proposition by combining the modality element, the AI element and the reading service element, question marks remain over Nanox’s ability to execute. There is potential in the integration, but the Israeli vendor’s inexperience, lack of established customers and sales networks, and its slender portfolio make the venture’s success a challenge. More concerning is the clear truth that other larger vendors are better positioned than Nanox to address and capitalise on the future of imaging diagnostics. If Nanox can make it work, and legitimately create value through the integrated company, more established vendors would also be able to create similar networks themselves. While they would not have Nanox’s first-mover advantage, the scale, connections, breadth of offering and market nous of a large modality vendor such as GE Healthcare, Philips, Siemens Healthineers or United Imaging, for example, would quickly leave Nanox under considerable pressure.

Ultimately, Nanox may have produced an innovative and potentially fruitful strategy with its latest acquisitions. Sadly, for the Israeli start-up, others look far more likely to harvest its rewards.

About Signify Premium Insights

This Insight is part of your subscription to Signify Premium Insights – Medical Imaging. The content is only available to companies that have subscribed to this paid-for service. To view other recent Premium Insights that are part of the service please click here