The purpose of this chapter is threefold.

Firstly, it is an honor to be included in this Festschrift, a recognition and celebration of the enormous contribution Professor Richard Baum has made to the field of nuclear medicine. I would like to write a few words as to why Richard’s contribution to nuclear medicine and theranostics has been so important from a commercial perspective, and why we need more innovators like him. Secondly, I want to review some of the reasons why, despite enormous potential, the field of nuclear medicine has not been as commercially successful as it could be, and the pitfalls we must address in order to deliver in the future. Finally, I’d like to highlight some of the areas that I am most excited about from a commercial perspective that will likely define the field over the next decade.

I’ve spent 20 years hunting money for imaging and nuclear medicine companies and over that time, through varying degrees of economic prosperity, I have received a great deal of candid feedback about how investors view the nuclear medicine industry. These perceptions offer provocative insights into our industry and I believe they are worth sharing, both for information and entertainment. Not all opinions will garner agreement.

4.1 Professor Richard Baum

Firstly, since this essay is delivered in honor and recognition of Richard, I would like to share a story about our esteemed colleague. In our field, there are few who have made such an immense contribution over such a long period of time, and even fewer who are such a force of nature. I first met Richard exactly ten years ago back in the early days of ImaginAb [1], where I was challenged by Richard to “get my act together” and do a first-in-human evaluation of an anti-PSMA antibody fragment (a minibody [2]) in an investigator-led study [3]. The truth is, I don’t think Richard likes biologics much and were it not for the fact that the prostate PSMAFootnote 1 imaging field was so nascent, I very much doubt I would have piqued his interest. But I am grateful he was curious because, thanks to Richard, we got some preliminary data that enabled us to raise a “Series A” venture capital financing for the company. Without this, the company would have simply failed to launch. I will come back to this again at the end of this essay.

It was a delicate situation. We had manufactured a first batch of (approximately) clinical-grade material, we didn’t have much of it to spare (and made at significant cost), but the company didn’t have a lot of financial resources either. After several months of harassment from Richard (I admit) we sent him a summary of the manufacturing package, noting that we had only a very basic radiolabeling protocol. After a week of contemplation, I received an email from Richard indicating that he would be keen to do a straightforward biodistribution study in patients with advanced metastatic prostate cancer. We packaged up the precious vials of DOTA-conjugated fragment and sent them to him, with enough additional material for the highly talented radiochemistry team at ZentralKlinik Bad Berka to undertake some basic process development, validation runs and a first patient study. We were expecting to image 8 patients in total with SPECT, labeled with 111In.

A few weeks went by, and we waited with great anticipation. Then one day a beautiful image appeared in my inbox. Then another. Then a more detailed biodistribution assessment and a rough dosimetry analysis. Then the 5th patient came. The images looked strange, a bit blurry (even by SPECT standards) and we were confused.

What had gone wrong?

Well, Richard had gone back to the first patient and substituted the indium for 177Lu and, based on the dosimetry had given an approximately 45 mCi/m2 dose of lutetium as a therapeutic dose. We were shocked, surprised, dumbfounded. It wasn’t supposed to be a therapeutic agent, it was supposed to be an imaging agent! I traveled from Los Angeles to Levi, FinlandFootnote 2 to have a meeting to discuss.

The meeting, which would forever change my view on the fundamental purpose of nuclear medicine, took place in a sauna at a resort a few hundred meters away from Santa’s Arctic village. It was November, and there was already plenty of snow on the ground—a far cry from the gentle winter climate of Southern California. I was so irritated and jet lagged, I didn’t really know what to say and I am sure my words blurted out in a jumbled mess. But Richard—as Richard often does—calibrated me on the facts of life.

He explained in his clear and typically direct manner, that his foremost mission is not to look at “interesting things” but to help his patients. He was encouraged (never went so far as to say “enthusiastic” I will note) by the imaging data and felt there was a chance to offer some benefit in the form of lutetium therapy. He explained that the patients had joined the study altruistically but that, in truth, there is little point in merely imaging disease. If patients are to benefit, then for a diagnostic there must be a corresponding therapeutic intervention, and without it—it is pointless. Pointless.

When you have spent a couple of years working on something with a pre-conceived idea of its value and purpose, to be re-calibrated in such a way is somewhat confronting. But what I observed in the years that followed is that I have met few clinician-scientists so committed and devoted to the patient. My personal lens of risk tolerance (or perhaps not) was simply an incomplete and perhaps even farcical viewpoint, and while it would clearly be unproductive for every investigator-led study to go off-piste, Richard always put the patient first.

The lesson I learned was simple—by all means do the experiment, but make sure that the purpose is maximized for the patient, not scientific curiosity. I wish I could say that this is an obvious and ingrained attribute of clinical research, but we all know that this is too often not the case, especially in nuclear medicine. It was a precious gift to me.

4.2 The Great Bexxar® Disaster

Moving on from sentimental rumination to business, the first reaction I typically get from investors in relation to anything nuclear medicine or radiopharmaceutical is a slight grimace and a slow exhalation of breath. A kinder investor audience will occasionally acknowledge that the technology is “nifty” but most just cut to the chase, namely that there are few good examples of high-quality nuclear medicine companies or commercially-successful products, even to this day. Investors prefer to see new ideas surrounded by relevant success stories and plenty of cash thrown at a given technology field. In investor parlance, they look for evidence that a company is “swimming where the water flows.” Nuclear medicine is complex and hard to sell, both technically and in terms of historical performance.

In 2014 I met with New York investment bankers with the intention of taking ImaginAb public. At the time, the tally of nuclear medicine disasters was fairly substantial. Lantheus had just failed an IPO attempt [4], Immunomedics’ share price was floundering about, Progenics wasn’t impressing investors and the reputation of our industry was mostly one of mediocrity from past failures like Corixa to the distracted implosion of IBA’s radiopharma business. While some of these companies have significantly reformed, even to the point of performing well, Bexxar and Zevalin® remain the epitome of our industry’s remarkable ability to produce clinically outstanding products that failed miserably in the commercial world.

Most readers of this festschrift know the reason why Bexxar failed and I don’t want this commentary to rehash old and mostly uninteresting history (although for those interested in an omprehensive written review, Luke Timmerman’s analysis of Bexxar is excellent [5]). Fundamentally, taking revenue away from the oncologist was a recipe for commercial disaster and, at the time, there was more money in delivering chemotherapy than there was selling a “one-shot” wonder drug, irrespective of patient benefit [6]. However, to simplify the Bexxar failure to revenue conflict and patient ownership is to be analytically superficial. The fact remains that despite the impressive performance of Bexxar in follicular lymphoma (an astonishing 75% complete response) [7], the clinical trials that could have catapulted a radiopharmaceutical into the front-line of cancer care were not particularly comprehensive, were lethargically marketed, and arguably did not compel the field to change the standard of care at the time.

From the Bexxar experience we can ascertain two important lessons about how to commercialize novel nuclear medicine products. The first lesson is that to deliver a successful product, there needs to be a great clinical dataset. Fast-forwarding a decade from Bexxar, one only has to consider the relative commercial success of the Algeta/Xofigo® journey to appreciate the reward for doing things properly. The ALSYMPCA trial [8] was a robustly designed trial, intended to produce the necessary level of evidence, and it certainly delivered. For 3.6 months of median survival benefit [9], Algeta created almost $3B of value for shareholders between initial partnering funding and ultimate acquisition by Bayer [10]. I personally don’t think anyone really believes that 223Ra-dicholoride is clinically any better than 153Sm or 89Sr (I certainly don’t), but Algeta did the hard work and snagged a significant prize. More recently we had the NETTER trial [11] for Lutathera® (177Lu-DOTATATE, AAA/Novartis) as another strong example of data-driven commercial outcomes for the field—and, of course, for patients. The point is, the growing call for robustly executed clinical trials is correct and necessary for the future of the field, for both clinical and commercial success.

4.3 “Us Versus Them”

The second insight from Bexxar relates to the turf battle between medical oncology and nuclear medicine. I have traveled extensively around the planet and observed the practice of nuclear medicine in many different clinical settings. Where there is a well-integrated team approach to treating cancer, experimental nuclear medicine (both diagnostic and therapeutic) thrives. But where there is an “us versus them” mentality, nuclear medicine remains relegated to the dungeon of the hospital and salvage patients waiting to exhale their last breath. It is a depressing and dreary place for otherwise outstanding science to exist.

This reality should encourage us to re-think the strategy for running clinical trials for nuclear medicine products, particularly therapeutic products. Does going head-head with standard care really sell the value proposition of nuclear medicine to an oncologist? Surely comparative trials that integrate standard care into the nuclear medicine treatment arm is a better way to go, especially given the therapeutic index typical of nuclear medicine approaches and the potential benefit in earlier stage patients? Considering some of the significant combination drug opportunities like androgen-deprivation therapy plus PSMA for prostate cancer [12], or combo radiation-checkpoint inhibitor studies [13], this would seem like a golden opportunity for our field. Especially since nuclear medicine therapies don’t tend to have overlapping toxicities [14] and imaging (i.e., patient selection, treatment response assessment) is often “built-in” for free [15]. The question is, will nuclear medicine ever really “play nice” with medical oncology?

Of course, much has changed in the past 5 years and now we have some decent success stories to talk about, at least at face value. I could be unkind and note with disdain how long it took for radioactive somatostatin analogs to broadly impact patient care—or whine a bit and note that we probably could have had a highly effective PSMA therapy a decade before Endocyte and PSMA-617 [16, 17], but we are making progress. Like many, I playfully mocked Stefano Buono (with no small amount of envy mixed in, I fully admit) when he moved into the 69th floor of the Empire State Building in New York and re-invented AAA [18]. But what a magnificent success! The reward comes from being courageous enough to take light-touch manufacturing processes and patchy clinical data out of the realm of academia into the harsh glare of the real world. Did intellectual property underpin AAA’s $4Bn valuation at the time it was acquired by Novartis? [19] Absolutely not—it was about doing the regulatory, manufacturing and clinical hard grind and finishing the job.

4.4 Intellectual Property: Who Cares?

Intellectual property (IP) is important but investors in our field probably worry too much about it. To be sure, we are probably going to see some interesting IP clashes in the near future, particularly around the PSMA small molecule programs [20]. There is also little doubt that IP infringement has hindered progress in the past. For example, we never had 18F-choline imaging in the United States, largely because of the deterrent effect of IP ownership uncertainty. While many men in Europe genuinely benefitted from choline, Americans missed out. Of course, PSMA imaging mostly makes choline and fluciclovine (Axumin) [21] imaging mostly obsolete [22], but it’s still disappointing to think of all the men that could have obtained some genuine benefit, not to mention the loss of a developmental incentive for the commercial 18F networks that would have almost certainly paved the way for a more flexible and capable manufacturing capacity, especially in the United States.

Although IP doesn’t particularly define our industry, especially given that it seems to take at least two decades for products to materialize (and thus composition of matter patents have expired anyhow), manufacturing and supply chain does define nuclear medicine, and this is where investors view our industry as especially challenging. There have been some interesting gambles—for example, when Bristol-Myers Squibb offloaded its imaging division to Avista Capital in 2008 [23], it was against the backdrop of a big pharma fear of the Cardiolite® patent cliff. Avista’s bet was basically that patent expiration wouldn’t particularly matter because supply chain and logistics trumped the importance of IP protection.

4.5 Imaging Is Not an Easy Business

It turned out that they were mostly right and although Lantheus is only now blossoming after its private equity hangover (with an incredibly low market capitalization relative to revenues, I might add), the Avista bet was a good one. However, market conditions aside, it is somewhat sobering to compare Lantheus’ failed 2014 IPO and slightly tepid (but ultimately successful) second-attempt in 2015 [24] with AAA’s highly respectable 2015 public offering [25]. Both companies offer medical imaging products but AAA had a superior financial profile because their investment thesis was focused on therapeutic medicine, not diagnostic imaging. In reality, Richard’s hot-tub pep talk is more than just about the patient, it also directly translates into value creation for shareholders.

Fundamentally, the commercial landscape of diagnostic nuclear medicine has undergone a dramatic shift. A decade or more ago, it was GE (via the Amersham acquisition), Siemens (PETNet), and IBA that were leading the charge in new nuclear medicine and PET tracers. Today none of those companies truly invest in new molecular imaging candidates and don’t contribute much to the momentum of diagnostic nuclear medicine, let alone do anything earth shattering in radionuclide therapy. At one point, companies like GE and Siemens were interested in new tracers because they thought it would help them to sell capital equipment. As a case in point, PETNet was established by CTI (now Siemens), precisely to fuel demand for scanners, the metaphor being one of gas stations (cyclotron sites) and cars (PET scanners).

Although the roll-out strategy of cyclotron networks and the success of FDG PET is something that Henry Ford would have immediately recognized as akin to roads and gas stations, the price erosion and extreme commoditization of FDG has made it challenging for new high-value tracers to get to market. It’s only recently with the advent of novel agents targeting PSMA (prostate) [26], CAIX (renal) [27], and immune cells (i.e., 18F-AraG [28], anti-CD8 [29], and anti-PD1/L1 [30] constructs) that the potential patient benefit above that of FDG is sufficient to command a decent price tag. Blue Earth Diagnostics is a case in point and did a laudable job with fluciclovine (Axumin®), with a dose commanding $4000+ a pop in the United States, at least until the point where bundled reimbursement potentially crushes pricing [31].

There is also no doubt that the commercial failure of amyloid imaging, despite three FDA approvals representing a combined investment of well over a billion dollars from GE, Avid (Lilly), and Bayer (Piramal) [32], has fundamentally tainted investor appetite to invest in new imaging agents as a stand-alone value proposition. Alzheimer’s imaging has clearly taught us that without a therapeutic intervention, an imaging agent is commercially useless [33]. With the possible dismissal of the “amyloid hypothesis” [34] likely follows the death of plaque imaging, despite arguable benefits to patient management [35]. Diagnostic nuclear medicine in cardiology and oncology plays a much more front-line role in guiding intervention, in the broadest sense, but the bar remains high for new stand-alone diagnostic products to truly show patient benefit. It is also at our peril that we ignore the various blood- and tissue-based diagnostic technologies that are making waves, and which certainly have the potential to compete in numerous applications [36].

4.6 Entrepreneurs Beware

What does this practically mean for the academic or entrepreneur (or, even better, academic-entrepreneur) who has a great idea for a new imaging tracer? Well, at a first pass, it probably means that there isn’t a straightforward business case. I encounter a lot of small stand-alone diagnostic imaging tracer companies that struggle to obtain meaningful investment (to do the necessary trials). I personally wouldn’t invest in a start-up that had an imaging tracer that wasn’t directly tied to a therapeutic intervention, no matter how interesting the clinical application.

I also don’t care much for made-up words like the term “theranostic” (and as an aside, I feel this term does our industry harm because it echos strongly of Theranos, the fraudulent blood diagnostics company [37] and we should endeavor to standardize a better term that investors reactively distrust less). However, I think that nothing in nuclear medicine is more important than the concept of theranostics. In fact, as we hurtle toward the era of personalized or “precision” medicine, it is truly our strategic advantage in comparison with other therapeutic modalities. It pains me to admit this because it means I personally wasted many years of my life trying to develop stand-alone diagnostic imaging products (in deference to Richard, I will admit my stupidity), but I firmly believe that in the long-term the strongest business model is where imaging is merely a cost of goods (COGS) of a therapy, and therefore should be as cheap and ubiquitous as possible.

Unfortunately, this isn’t just history. New imaging tracers are still being developed to tackle a mostly theoretical unmet need. While I am as enamored as anyone with the beautiful images from FAPI [38], the world simply does not need another FDG and this new technology needs to developed to be more than “the next FDG”. Indeed, one of the biggest business case impediment to developing a new broad-use imaging agent is precisely the power and beauty of FDG. It’s slightly unfortunate that our first great PET imaging may possibly be the best we’ll ever have. FDG has big shoes to fill and unless the concordant therapeutic benefit of new strategies can be demonstrated, we will see plenty of new (however appealing) imaging agents fail to transform clinical care, and therefore add to the list of commercial failures by extension.

4.7 Build It and They Will Come

A “theranostic” strategy, the commitment to running proper clinical trials and integrating with standard care could mean a very bright future for our field. Judging by the packed attendance at PSMA and NET conference presentations, it’s clear we have plenty of attention from referral physicians. Richard has played a pivotal role in the creation of this incredible opportunity, including the momentum and enthusiasm we enjoy today. For the first time in a long time, it’s exciting to be in nuclear medicine, but there are still plenty of ways to mess it up if we are not careful.

Aside from the marginally useful clinical studies we have a tendency to run, “big pharma” has traditionally shunned nuclear medicine because of its manufacturing and supply chain complexity. Our products are complicated to make, are “melting ice cubes” and the just-in-time logistics of shipping a product anywhere in the world, every day, requires a heroic effort. Unfortunately, the supply chains that underpin our industry are not as strong as they need to be for future success.

The relentless commoditization of FDG means that our 18F cyclotron networks are generally in poor shape and need a significant level of investment to bring them up to the standard required to deliver multi-product capability, particularly considering new manufacturing regulations in the United States and EU. This will be a significant investment headwind for new tracers, particularly 18F-based PSMA tracers [39, 40] and notwithstanding the generally impressive commercial success of Axumin in the United States, roll-out was not particularly fast. For those with a penchant for 68Ga, it’s fair to say that most generator vendors were caught short [41] and have unfortunately tarnished the reputation of gallium with investors.

4.8 A Fragile Supply Chain

On the therapeutic side, I believe we are on the verge of our first supply chain crunch for no carrier added (NCA) 177Lu. By the way, as an industry, we should only be using NCA lutetium if we are going to scale an industry based on responsible waste management and a tenable environmental profile. Although there is some optimization possible and there is talk of re-processing “raw” chemical-grade 177Lu from various high-output reactor sources around the globe, scaling the supply chain remains a non-trivial exercise. There are already some early signs of the “crunch”, with vendors starting to sell preferential and guaranteed access to production capacity [42], which I argue wouldn’t happen if production was truly near-term scalable. Otherwise why diminish the value (and valuation) of the supply chain in such a simplistic way? Although there is seemingly a plethora of new NCA lutetium production projects around the globe, only a small number are really serious and they are probably at risk of acquisition, possibly to the exclusion of the overall growth of the industry. If companies cannot access at least two or three credible, stable suppliers of isotope, then there will be insufficient resilience of supply to support product development and commercialization.

The issue goes considerably beyond lutetium. There are several promising 131I products that are approved or close to market (Progenics [43], Y-mAbs [44]), but despite cheap and plentiful availability of iodine, manufacturing capacity for iodinated drug product remains extremely limited in virtually every territory. With a resurgence of interest in therapeutic nuclear medicine, driven by lutetium and the allure of alpha emitters, it’s reasonable to assume that 131I (and perhaps other useful flavors of iodine) [45] is going to see a renaissance too. Frankly, it’s a very good isotope and worth a second life, even if success will mean investing in production and clinical infrastructure.

It almost doesn’t matter what the therapeutic isotope is, the specialist manufacturing infrastructure is patchy. Capacity for clinical trials and even early commercialization of the odd orphan drug is available, but scale-up manufacturing essentially remains an unsolved problem in most parts of the world. I believe that the most powerful part of our ecosystem could end up becoming firms like Curium (the merger of IBA Molecular and Mallinckrodt’s isotope business) [46] and Cardinal Health, and we are seeing considerable industry consolidation around manufacturing and production networks at present. There is a plausible scenario whereby a few dominant companies define the economics and market access reality of our industry. This in turn could have a suppressive effect on new product innovation as there would be little incentive to invest in early-stage clinical translation because it doesn’t generate meaningful revenue and it’s not evident that the larger supply chains are interested in investing in early-stage assets. New contract development and manufacturing organizations (CDMOs), like Evergreen Theragnostics [47], are badly needed to build this future capacity in a cost-effective way.

4.9 Future Frontiers

I am not particularly gifted at gazing into the crystal ball, I will save that task to Richard. Our field is still a moving target and we should persist in addressing the systemic risks of our industry. It would be hard to remain motivated and optimistic about new product development if there were not clear pathways forward to market. That said, there is always the temptation to skip past what we already have on our plate and play with the next shiny toy. Because our industry is so academically driven, we essentially have zero attention span and every time I attend the major scientific conferences I am reminded that part of the reason why we don’t have great products impacting healthcare is because academics do the cool research, write the paper, run their clinical proof-of-concept (for another nice paper) and then move on. For the most part, new ideas don’t end up in commercial ventures.

Although I am intrigued about the potential for alpha emitters (particularly 225Ac, 211At, and possibly 212Pb) there is truthfully very little actual innovation going on here in my opinion. Putting a different isotope onto the same old targeting agents and declaring success isn’t going to enable the field of alpha therapeutics [48]. We basically never did any meaningful prospective MTD/dose optimization studies with 177Lu peptide therapies [49] and it doesn’t look like anyone is really doing it with alpha emitters either, which is concerning for both the science and the patient. In the case of isotopes with complex decay chains like 225Ac [50] or 227Th (if one must) [51] there is so much more fundamental research and long-term follow-up to be done before we can robustly turn to our colleagues in mainstream oncology and say “we have something you should use.” The fundamental radiobiology really isn’t being done at all and it’s going to hurt us at some point.

There is also a lot of mythology in our field. One of the great myths is that if we make our products more “drug-like” (i.e., use alphas) that mainstream pharma will finally buy into radiopharmaceuticals in a big way. Actually, there is no real basis for this assertion and Novartis’ acquisition of AAA and Endocyte should be considered to be evidence of this misconception. What matters is cost-effective manufacturing and meaningful clinical data, not the seductive characteristics of a particular radionuclide. In my opinion, diversifying the manufacturing capability of our industry is vital, not just for therapeutics but for diagnostic imaging products too. The use of isotopes like 89Zr and 64Cu (to a limited extent) has the potential to alleviate the need to run large-scale networks of cyclotron production sites for imaging, which would have a big impact on production economics and patient access.

Finally, as I have already alluded, I think the most exciting future for the oncology-centric part of nuclear medicine, is immuno-oncology. On the imaging side, we have a growing number of PET tracers that provide an incredible window into the immune system and there is some evidence that mainstream pharma is paying attention [52]. On the therapeutic side, there is probably no therapeutic modality that is more synergistic with immuno-oncology drugs than targeted radiation, although the combination with external beam radiation is exciting too [53, 54]. If one considers the fundamental radiobiology, one could even argue that using targeted radiation to invoke an immune response in a highly complex and heterogeneous tumor micro-environment may be better accomplished with beta-rather than alpha-emitters (heresy, I know). We don’t have to wait until tomorrow to explore this potential and I firmly believe it will become our industry’s finest hour, providing that we can develop the analytical methods and clinical software applications to quantitate what is actually happening, and how to optimally cycle therapy for both toxicity and patient benefit.

4.10 Concluding Remarks

Buoyed by new optimism and the recognition that our industry has finally come of age, it’s time to move out from our “spotty teenager” phase and into the relative grace of adulthood. With industry and academia working together more effectively than ever, and academics increasingly demonstrating risk appetite for new commercial ventures, perhaps we will soon see a time where radiopharmaceuticals, and not CAR-Ts or gene editing tools, command the headline at mainstream biotech investor conferences.

As I have asserted, we need to work with mainstream oncology and integrate with standard care. We need to run scaled, prospective clinical studies that use meaningful comparators, not quasi-standards that don’t reflect best practice [55]. Our manufacturing and supply chain needs to become more durable, flexible, and diversified.

However, my personal wish—offered with thanks and deference to our friend and colleague Richard—is that we also continue to take appropriate and patient-centric translational risks. Although the regulatory environment for nuclear medicine is not getting easier or less onerous, we still participate in a field that demands enormous cross-functional technical discipline to deliver. With this discipline and the elegance of what “theranostics” can achieve, we should be able to take measured risk, develop products in a more streamlined way, and deliver patient benefit faster than other fields of medicine. Commercial success, particularly for new startups, will utterly depend on this dynamic. Nuclear medicine has the potential to evolve from the “wild west” of the last couple of decades to a much more polished modality, but “cowboys” are still wanted, worthy of our finest - Richard Baum. Congratulations on your transformational and disruptive career!