Goldman's Top Takeaways From Its Semiconductor Conference As Tech Execs Focused On AI

 Last week, Goldman Sachs hosted the 2nd Annual Global Semiconductor Conference in New York, where they gained valuable insight into where the industry is headed from management and IR teams from the semiconductor device, equipment, and materials companies. Conversations were also held with Todd Fisher, the person the Biden-Harris Administration appointed to lead the CHIPS for America offices. 

Goldman's Toshiya Hari said there was a lot of focus on artificial intelligence from participants, including Intel, Marvell, Micron, Renesas, and Advantest. Management teams of these companies overwhelmingly believe AI will be 'long-term' growth drivers, though some said it might take time for the growth to be realized.  

There were signs from Intel and Micron that the PC bust cycle might be stabilizing. As well as signs the memory industry is finally "bottoming." 

Fisher provided more clarity on the Biden administration's long-term goals of building out America's domestic chip production while pointing out there is "no bias either way in the treatment of a domestic or international applicant as the goal of the program is to encourage companies to invest in R&D and for IP to reside in the United States." 

Here's Goldman's Hari summary of the top ten takeaways from the chip conference: 

1) Focus on AI: 

There was an immense focus on AI throughout our conference with Intel, Marvell, Micron, Renesas, and Advantest, in particular, speaking to the near- and long-term opportunity set associated with this growing theme. Intel highlighted how Sapphire Rapids (i.e. 4th generation Xeon scalable processors based on Intel 7 technology) is well-suited for AI workloads (note Nvidia selected Sapphire Rapids as the standard server CPU in its DGX H100 system last year), while management also shared that its pipeline for Gaudi (i.e. Habana's training and inference accelerator) had increased ~2.5x in the preceding 90-day period. Marvell reiterated what it had disclosed the prior week on its earnings call — namely, that optical DSPs and custom compute processors are expected to lead to a more than doubling of AI revenue in FY2024 and FY2025 from a base of ~$200mn in FY2023. Micron stated that although AI revenue is difficult to quantify and it currently makes up a small percentage of total revenue, they see AI as a significant long-term growth driver given the implications for content growth. While there is a range, Micron believes AI servers can embed 8x the amount of DRAM and 3x the amount of NAND compared to a traditional server. Renesas highlighted the medium- to long-term growth potential in MCUs, particularly at the edge (i.e. multi-billion dollar SAM), their recent acquisition of Reality AI, a predictive AI company, that will augment its MCU capabilities particularly across industrial applications (e.g. HVAC), as well as its ongoing investments in CXL memory accelerators. For Advantest, while HPC/AI-related demand is unlikely to move the needle on CY2023 tester demand, per management, the company sees HPC/AI as a medium- to long-term growth driver given a) the expected increase in transistor count, b) the potential increase in test intensity as the industry accelerates the adoption of advanced packaging, and c) the company's confidence in defending its dominant share position in this market segment.

2) Signs of stabilization in the PC market: 

Signs of stabilization are emerging in the PC market with Intel raising the mid-point of its 2Q revenue outlook from $12.0bn to $12.25bn (+5% qoq, -20% yoy) based on strong linearity in Client Computing (i.e. PC) and Data Center and AI so far in the quarter and Micron reiterating its expectation for customer inventory in PCs (and smartphones) to be at or near normal levels exiting the CY2Q. While sell-in of components in CY2H and beyond will depend on PC sell-through, we expect, at a minimum, the under-shipping of components relative to end-demand that has persisted over the past ~9 months to subside soon.

3) Memory fundamentals bottoming: 

While Micron's disclosure that the recent ruling by the Cyberspace Administration of China (CAC) could have a high-single-digit (%) impact on total revenue, up from the low- to high-single-digit (%) range provided by management on 5/22, weighed on the stock last week, our constructive view on the Memory cycle predicated on demand stabilization and supply-side discipline (i.e. capex and production cuts) remains intact. Between DRAM and NAND, we continue to expect a sharper and more sustained recovery in DRAM given relative inventory levels (i.e. DRAM < NAND) and relative industry consolidation as measured by HHI (i.e. DRAM > NAND). In NAND, we fear that suppliers with relatively weak balance sheets could re-accelerate bit production once pricing has recovered to above cash cost.

4) Benign pricing in analog/MCU/power semis:

In broad-based MCU, analog and power semis, Microchip and Infineon, in contrast to growing investor skepticism, pointed to stable industry trends. Microchip reaffirmed its June quarter (+2-3% qoq) and September quarter (unlikely to be down qoq) revenue outlook, while Infineon reiterated its confidence in its auto semis growth outlook with underlying unit demand still solid in Europe/US. On pricing, Infineon stated that pricing remains resilient across all divisions, and is even increasing in certain pockets where demand is strong. Similarly, Microchip spoke to stable near-term pricing and shared its view that industry pricing is likely to be less deflationary going forward than in the past given higher capital intensity across mature process nodes.

5) TEL presents bullish CY2024 WFE market outlook: 

While the majority of Wafer Fab Equipment (WFE) suppliers have yet to comment on CY2024, Tokyo Electron (TEL) reiterated its view that the WFE market in CY2024 could recover to a level similar to CY2022 (which implies a ~25% yoy increase), driven by a data center upgrade cycle and a recovery in Memory spending following this year's sharp inventory adjustment. Note that our own expectations for the WFE market in CY2024 are more subdued at +7% yoy based on a double-digit yoy increase in Memory and a stable outlook in advanced Logic/Foundry, partially offset by a decline across mature/specialty nodes.

6) Gate-All-Around to drive advanced Logic/Foundry spend: 

Applied Materials, ASML, ASM International, and Tokyo Electron all highlighted Gate-All-Around (GAA) as a potential driver of higher spending in advanced Logic/Foundry over the coming several years. ASM International highlighted that it will begin to receive GAA orders in 4Q23 and that it expects growth in its Epitaxy business to be catalyzed by the transition to GAA. Applied Materials, on its recent earnings call, stated that the GAA inflection will create an incremental opportunity of ~$1bn for every 100k wafer starts of capacity and that it expects to gain 5% of transistor market share in the transition from FinFET to GAA, particularly in product areas including Epitaxy and Selective Removal, in our view.

7) Constructive long-term outlook on mature node capital investments:

 Applied Materials reiterated that its ICAPS (IoT, Communications, Automotive, Power and Sensors) business is on track to grow in CY2023 at a faster pace than in CY2022 given strength across China, Japan, Europe, and the US While we expect capital spending across mature/specialty nodes to remain cyclical, we subscribe to the view that capital intensity in the trailing-edge will stay elevated vis-a-vis the past 5-10 years as the used equipment market the IDMs and foundry suppliers used to leverage has since declined in size. Note TEL stated that they expect WFE demand associated with mature process nodes could reach ~$50bn by CY2030, up from ~$30bn in CY2023, while ASML addressed skepticism surrounding spending on mature/specialty nodes in China by sharing that ordered lithography tools are being installed in cleanrooms (rather than only being ordered for strategic/geopolitical purposes and stored).

8) Industry wafer starts to recover in 2H: 

Entegris reaffirmed its CY2023 market outlook — specifically, a mid-teens (%) yoy decline in MSI and a ~20% yoy decline in industry capex. That said, the company expects a modest recovery in 2H23 driven by advanced Logic/Foundry on growth in AI and the introduction of new consumer electronics products. Management remains confident in its ability to deliver consistent outgrowth — 6-7% points this year — as customers' execute to their respective technology transitions (e.g. Gate-All-Around) and in turn consume more of Entegris' products on a per-wafer basis.

9) Near-term caution on wafer volumes but ASP outlook intact: 

SUMCO shared a relatively cautious outlook for its silicon wafer business as the ongoing inventory correction in Memory is likely to drive a hoh decline in shipments in 2H. On a positive note, management stated that wafer pricing continues to track largely in-line with what had been agreed in LTAs and that the current expectation is for wafer pricing to increase ~10% yoy in CY2024.

10) CHIPS Act: 

from the CHIPS for America program, we hosted Todd Fisher who had spent 30 years in the finance and investment industry, including nearly 25 years at KKR & Co. Inc., prior to joining the Department of Commerce in 2021. Related to the CHIPS Act, Mr. Fisher shared the US Government's long-term goals, including a) at a high level, the pursuit of economic and national security, and at a micro level, b) the construction of at least two new leading-edge Logic/Foundry eco-systems in the US by the end of the decade, as well as c) the creation of a resilient supply chain as it pertains to mature process nodes and specialty technologies. Interestingly, Mr. Fisher noted that there is no bias either way in treatment of a domestic or international applicant as the goal of the program is to encourage companies to invest in R&D and for IP to reside in the United States. In his concluding remarks, Mr. Fisher summarized the six criteria under which applications are evaluated: 1) impact to economic and national security (the most significant), 2) financial viability, 3) commercial viability (including potential long-term implications for industry supply/demand), 4) technical feasibility, 5) workforce, and 6) broader impacts (with a significant discussion around R&D).

The explosion of interest in AI might be a growth driver of the semiconductor sector in two ways: building demand for innovative technologies and increasing chip demand. 

https://www.zerohedge.com/markets/here-are-goldmans-top-takeaways-its-semiconductor-conference-tech-execs-focused-ai

Asco 2023 – Commands leaves the door open for Geron

 The recently intensifying battle in myelodysplastic syndromes between Bristol Myers Squibb’s Reblozyl and Geron’s imetelstat appears to have turned in favour of the latter’s underdog: presentation of the front-line Commands trial at Asco has confirmed Reblozyl’s limited activity in patients without ringed sideroblasts.

A pre-Asco press briefing put a positive spin on the data, but the results speak for themselves. According to Commands’ primary endpoint there was no benefit at all in ringed sideroblast-negative subjects, who comprise some two thirds of this MDS population. Crucially, Geron’s Imerge trial did suggest that imetelstat was active in these patients.

There are further distinctions, most importantly that Commands is a first-line MDS trial, pitting Reblozyl against erythropoiesis-stimulating agents (ESAs), while Imerge tested imetelstat in patients progressed on or ineligible for ESAs.

Reblozyl is already approved for post-ESA use in low-risk MDS, though only in ringed sideroblast (RS) positive patients – a fact that apparently does not preclude some off-label prescribing in RS-negative MDS. Full data from Imerge will be presented alongside the Commands results at an Asco session on 2 June.

Paradigm shift

Presenting data to the press this week Dr Guillermo Garcia-Manero, from the MD Anderson Cancer Center, called Commands a “paradigm shift in the treatment of low-risk MDS-associated anaemia”, and a result thanks to which Reblozyl could replace ESAs as a new front-line option.

The basis for this enthusiasm was that Reblozyl was nearly twice as likely to result in transfusion independence with haemoglobin increase than epoetin alfa. 58.5% of Reblozyl recipients achieved this at 12 weeks, Commands’ primary endpoint, versus 31.2% of patients given epoetin alfa (p<0.0001).

However, this all-comers effect was clearly driven by RS-positive patients. Those who were RS-negative derived no added benefit, with 12-week transfusion independence of 41.0% for Reblozyl versus 46.3% for epoetin alfa.

Source: Dr Guillermo Garcia-Manero & Asco.

Despite this, Garcia-Manero argued that Reblozyl should be used irrespective of RS status. Commands did not reflect the real-world population because over 60% of enrolled patients were RS-positive, and the trial was “not powered to see a big difference ... in the RS-negative context”, he told the Asco press briefing.

Furthermore, he pointed to duration of transfusion independence, a secondary Commands endpoint that he suggested was as important as response itself. Here there was a strong numerical benefit favouring Reblozyl versus epoetin alfa among all-comer, RS-positive and RS-negative patients.

Commands prompt

The importance of Commands is twofold. Firstly it will likely result in Reblozyl getting a front-line label in low-risk MDS – though clearly how broad this is has yet to be determined. Secondly, depending on the extent to which the Bristol drug replaces front-line ESAs, it will change the second-line treatment landscape.

However, as long as Reblozyl is not indicated for RS-negative MDS the second-line imetelstat opportunity, which always looked likely to be limited to this population, should remain intact. Moreover, though the Geron project has not been tested in post-Reblozyl MDS, a recent KOL survey cited by B Riley analysts suggested that prescribers would use it in 55% of such patients.

Evaluate Pharma sellside consensus shows forecast 2028 revenues of $2.5bn for Reblozyl and $849m for imetelstat. Though each bank will make different assumptions about the chances of front-line use and/or an RS-agnostic label, if Reblozyl remains limited to RS-positives the peak sales forecasts could take a hit.

According to Garcia-Manero Reblozyl’s response duration, relative safety and ease of SC administration every three weeks means that the drug “likely will become the standard of care for the majority of [low-risk] patients, regardless of whether they are RS-positive or negative”. Whether this in fact happens depends on prescribers and the FDA.

https://www.evaluate.com/vantage/articles/events/conferences-trial-results/asco-2023-commands-leaves-door-open-geron

Musk: Target can expect shareholder lawsuits soon as retailer's woes mount

 Billionaire entrepreneur Elon Musk says it is just a matter of time before Target Corp. is hit with legal action from investors as the retailer continues to shed value amid controversy over its Pride merchandising decisions.

In response to the news that JPMorgan Chase & Co. downgraded Target stock from "overweight" to "neutral" last week, Musk tweeted, "Won’t be long before there are class-action lawsuits by shareholders against the company and board of directors for destruction of shareholder value."

Tesla CEO Elon Musk predicts Target could be hit with shareholder lawsuits as the retailer continues to struggle amid controversy surrounding its Pride merchandising decisions. (Reuters/Aly Song//File / Reuters)

Musk, the CEO of Tesla, Twitter and SpaceX and the world's richest person, was replying to conservative commentator Charlie Kirk reporting on Target's first of two stock downgrades over the past week.

Kirk noted Target stock is on "its longest losing streak in 23 years" and added, "Happy Pride Month Target!"

Target has been hit with backlash from both the right and left over its handling of LGBTQ merchandising this year.

The company irked conservatives last month over its LGBTQ merchandise displays as it prepared for Pride month, which many viewed as excessive The retailer, which has supported LGBTQ Pride for years, this season offered merchandise that included female-style swimsuits that have the option to "tuck" male genitalia.

JPMorgan Chase & Co. downgraded Target stock from "overweight" to "neutral" last week, with analysts citing the possibility of a decline in sales due to consumers pulling back spending amid persistent inflation.

Ticker	Security	Last	Change	Change %
TGT	TARGET CORP.	130.52	-2.69	-2.02%
Powered By

Target then angered liberals by relocating and removing some LGBTQ items in an effort to tamp down customer "outrage," as first reported by FOX News Digital.

Since the backlash began, Target's market value has fallen more than $13 billion to $60.24 billion as of Monday's closing price.

https://www.foxbusiness.com/markets/elon-musk-target-shareholder-lawsuits-soon-retailers-woes-mount

Game-Changing Alzheimer’s Research: The Latest on Biomarkers

 The field of neurodegenerative dementias, particularly Alzheimer’s disease (AD), has been revolutionized by the development of imaging and cerebrospinal fluid biomarkers and is on the brink of a new development: emerging plasma biomarkers. Research now recognizes the relationship between the cognitive-behavioral syndromic diagnosis (ie, the illness) and the etiologic diagnosis (the disease) — and the need to consider each separately when developing a diagnostic formulation. The National Institute on Aging and Alzheimer’s Association Research Framework uses the amyloid, tau, and neurodegeneration system to define AD biologically in living patients. Here is an overview of the framework, which requires biomarker evidence of amyloid plaques (amyloid positivity) and neurofibrillary tangles (tau positivity), with evidence of neurodegeneration (neurodegeneration positivity) to support the diagnosis.

The Diagnostic Approach for Symptomatic Patients

The differential diagnosis in symptomatic patients with mild cognitive impairment (MCI), mild behavioral impairment, or dementia is broad and includes multiple neurodegenerative diseases (eg, AD, frontotemporal lobar degeneration, dementia with Lewy bodies, argyrophilic grain disease, hippocampal sclerosis); vascular ischemic brain injury (eg, stroke); tumors; infectious, inflammatory, paraneoplastic, or demyelinating diseases; trauma; hydrocephalus, toxic/metabolic insults; and other rare diseases. The patient’s clinical syndrome narrows the differential diagnosis.

Once the clinician has a prioritized differential diagnosis of the brain disease or condition that is probably causing or contributing to the patient’s signs and symptoms, they can then select appropriate assessments and tests, typically starting with a laboratory panel and brain MRI. Strong evidence backed by practice recommendations also supports the use of fluorodeoxyglucose PET as a marker of functional brain abnormalities associated with dementia. Although molecular biomarkers are typically considered at the later stage of the clinical workup, the anticipated future availability of plasma biomarkers will probably change the timing of molecular biomarker assessment in patients with suspected cognitive impairment owing to AD.

Molecular PET Biomarkers

Three PET tracers approved by the US Food and Drug Administration (FDA) for the detection of cerebral amyloid plaques have high sensitivity (89%-98%) and specificity (88%-100%) compared with autopsy, the gold standard diagnostic tool. However, these scans are costly and are not reimbursed by Medicare and Medicaid. Because all amyloid PET scans are covered by the Veterans Administration, this test is more readily accessible for patients receiving VA benefits.

The appropriate-use criteria developed by the Amyloid Imaging Task Force recommends amyloid PET for patients with persistent or progressive MCI or dementia. In such patients, a negative amyloid PET scan would strongly weigh against AD, supporting a differential diagnosis of other etiologies. Although a positive amyloid PET scan in patients with MCI or dementia indicates the presence of amyloid plaques, it does not necessarily confirm AD as the cause. Cerebral amyloid plaques may coexist with other pathologies and increase with age, even in cognitively normal individuals.

The IDEAS study looked at the clinical utility of amyloid PET in a real-world dementia specialist setting. In the study, dementia subspecialists documented their presumed etiologic diagnosis (and level of confidence) before and after amyloid PET. Of the 11,409 patients who completed the study, the etiologic diagnosis changed from AD to non-AD in just over 25% of cases and from non-AD to AD in 10.5%. Clinical management changed in about 60% of patients with MCI and 63.5% of patients with dementia.

In May 2020, the FDA approved flortaucipir F-18, the first diagnostic tau radiotracer for use with PET to estimate the density and distribution of aggregated tau neurofibrillary tangles in adults with cognitive impairment undergoing evaluation for AD. Regulatory approval of flortaucipir F-18 was based on findings from two clinical trials of terminally ill patients who were followed to autopsy. The studies included patients with a spectrum of clinically diagnosed dementias and those with normal cognition. The primary outcome of the studies was accurate visual interpretation of the images in detecting advanced AD tau neurofibrillary tangle pathology (Braak stage V or VI tau pathology). Sensitivity of five trained readers ranged from 68% to 86%, and specificity ranged from 63% to 100%; interrater agreement was 0.87. Tau PET is not yet reimbursed and is therefore not yet readily available in the clinical setting. Moreover, appropriate use criteria have not yet been published.

Molecular Fluid Biomarkers

Cerebrospinal (CSF) fluid analysis is currently the most readily available and reimbursed test to aid in diagnosing AD, with appropriate-use criteria for patients with suspected AD. CSF biomarkers for AD are useful in cognitively impaired patients when the etiologic diagnosis is equivocal, there is only an intermediate level of diagnostic confidence, or there is very high confidence in the etiologic diagnosis. Testing for CSF biomarkers is also recommended for patients at very early clinical stages (eg, early MCI) or with atypical clinical presentations.

A decreased concentration of amyloid-beta 42 in CSF is a marker of amyloid neuritic plaques in the brain. An increased concentration of total tau in CSF reflects injury to neurons, and an increased concentration of specific isoforms of hyperphosphorylated tau reflects neurofibrillary tangles. Presently, the ratios of t-tau to amyloid-beta 42, amyloid-beta 42 to amyloid-beta 40, and phosphorylated-tau 181 to amyloid-beta 42 are the best-performing markers of AD neuropathologic changes and are more accurate than assessing individual biomarkers. These CSF biomarkers of AD have been validated against autopsy, and ratio values of CSF amyloid-beta 42 have been further validated against amyloid PET, with overall sensitivity and specificity of approximately 90% and 84%, respectively.

Some of the most exciting recent advances in AD center around the measurement of these proteins and others in plasma. Appropriate-use criteria for plasma biomarkers in the evaluation of patients with cognitive impairment were published in 2022. In addition to their use in clinical trials, these criteria cautiously recommend using these biomarkers in specialized memory clinics in the diagnostic workup of patients with cognitive symptoms, along with confirmatory CSF markers or PET. Additional data are needed before plasma biomarkers of AD are used as standalone diagnostic markers or considered in the primary care setting.

We have made remarkable progress toward more precise molecular diagnosis of brain diseases underlying cognitive impairment and dementia. Ongoing efforts to evaluate the utility of these measures in clinical practice include the need to increase diversity of patients and providers. Ultimately, the tremendous progress in molecular biomarkers for the diseases causing dementia will help the field work toward our common goal of early and accurate diagnosis, better management, and hope for people living with these diseases.

Bradford C. Dickerson, MD, MMSc

Professor, Department of Neurology, Harvard Medical School; Director, Frontotemporal Disorders Unit, Department of Neurology, Massachusetts General Hospital, Boston, Massachusetts

Disclosure: Bradford C. Dickerson, MD, MMSc, has disclosed no relevant financial relationships.

https://www.medscape.com/viewarticle/992555

Sucralose Damages DNA, Linked to Leaky Gut: Study

 A new study reveals health concerns about the sugar substitute sucralose so alarming that researchers said people should stop eating it and the government should regulate it more.

Sucralose is sold under the brand name Splenda and is also used as an ingredient in packaged foods and beverages.

The findings were published this week in the Journal of Toxicology and Environmental Health, Part B. The researchers conducted a series of laboratory experiments exposing human blood cells and gut tissue to sucralose-6-acetate. The findings build on previous research that linked sucralose to gut health problems.

The researchers found that sucralose causes DNA to break apart, putting people at risk for disease. They also linked sucralose to leaky gut syndrome, which means the lining of the intestines are worn down and become permeable. Symptoms are a burning sensation, painful digestion, diarrhea, gas, and bloating.

When a substance damages DNA, it is called genotoxic. Researchers have found that eating sucralose results in the body producing a substance called sucralose-6-acetate, which the new study now shows is genotoxic. The researchers also found sucralose-6-acetate in trace amounts in off-the-shelf products that are so high, they would exceed the safety levels currently allowed in Europe.

"It's time to revisit the safety and regulatory status of sucralose because the evidence is mounting that it carries significant risks. If nothing else, I encourage people to avoid products containing sucralose," said researcher Susan Schiffman, PhD, adjunct professor of biomedical engineering at North Carolina State University, in a statement. "It's something you should not be eating." 

The FDA says sucralose is safe, describing it as 600 times sweeter than table sugar and used in "baked goods, beverages, chewing gum, gelatins, and frozen dairy desserts."

"To determine the safety of sucralose, the FDA reviewed more than 110 studies designed to identify possible toxic effects, including studies on the reproductive and nervous systems, carcinogenicity, and metabolism," the agency explained on its website. "The FDA also reviewed human clinical trials to address metabolism and effects on patients with diabetes."

Sources

North Carolina State University: "Chemical Found in Common Sweetener Damages DNA."

Cleveland Clinic: "Leaky Gut Syndrome."

Journal of Toxicology and Environmental Health, Part B: "Toxicological and pharmacokinetic properties of sucralose-6-acetate and its parent sucralose: in vitro screening assays."

FDA: "Aspartame and Other Sweeteners in Food."

https://www.medscape.com/viewarticle/992667

FDA OKs Injectafer for Iron Deficiency Anemia in Heart Failure

 The US Food and Drug Administration (FDA) has expanded the indication for ferric carboxymaltose injection (Injectafer, Daiichi Sankyo/American Regent) to include treatment of iron deficiency in adults with New York Heart Association (NYHA) class II/III heart failure (HF).

"This new indication for Injectafer marks the first and only FDA approval of an intravenous iron replacement therapy for adult patients with heart failure," Ravi Tayi, MD, MPH, chief medical officer at American Regent, said in a news release.

Ferric carboxymaltose injection is also indicated for the treatment of iron deficiency anemia in adults and children as young as 1 year of age who have either intolerance or an unsatisfactory response to oral iron, and in adult patients who have nondialysis dependent chronic kidney disease.

The new indication in HF was supported by data from the CONFIRM-HF randomized controlled trial that evaluated the efficacy and safety of ferric carboxymaltose injection in adults with chronic HF and iron deficiency.

In the study, results showed that treatment with ferric carboxymaltose injection significantly improved exercise capacity compared with placebo in iron-deficient patients with HF.  

No new safety signals emerged. The most common treatment emergent adverse events were headache, nausea, hypertension, injection site reactions, hypophosphatemia, and dizziness.

According to the company, ferric carboxymaltose injection has been studied in more than 40 clinical trials that included over 8800 patients worldwide and has been approved in 86 countries.

https://www.medscape.com/viewarticle/992793

Transplant Centers Often Skip the Top Spot on the Kidney Waitlist

 Welcome to Impact Factor, your weekly dose of commentary on a new medical study. I'm Dr F. Perry Wilson of the Yale School of Medicine.

The idea of rationing medical care is anathema to most doctors. Sure, we acknowledge that the realities of healthcare costs and insurance companies might limit our options, but there is always a sense that when something is truly, truly needed, we can get it done.

Except in one very particular situation, a situation where rationing of care is the norm. That situation? Organ transplantation.

There is no way around this: More patients need organ transplants than there are organs available to transplant. It is cold, hard arithmetic. No amount of negotiating with an insurance company or engaging in prior authorization can change that.

As a kidney doctor, this issue is close to my heart. There are around 100,000 people on the kidney transplant waiting list in the US, with 3000 new patients being added per month. There are only 25,000 kidney transplants per year. And each year, around 5000 people die while waiting for a transplant.

A world of scarcity, like the world of kidney transplant, is ripe for bias at best and abuse at worst. It is in part for that reason that the Kidney Allocation System exists. It answers the cold, hard arithmetic of transplant scarcity with the cold, hard arithmetic of a computer algorithm, ranking individuals on the waitlist on a variety of factors to ensure that those who will benefit most from a transplant get it first.

But, according to a new study in JAMA Network Open, some centers seem to be treating this list as more of a suggestion than a rule. And that could be a real problem.

This area is a bit complex but I'll try to break it down into what you need to know. There are 56 organ procurement organizations (OPOs) in the United States. These are nonprofits with the responsibility to recover organs from deceased donors in their area.

 

Each of those OPOs maintains a ranked list of those waiting for a kidney transplant. Depending on the OPO, the list may range from a couple hundred people to a couple thousand, but one thing is the same, no matter what: If you are at the top of the list, you should be the next to get a transplant.

Most OPOs have multiple transplant centers in them, and each center is going to prioritize its own patients. If a Yale patient is #1 on the list and a kidney offer comes in, it would be a good idea for us to accept, because if we reject the offer, the organ may go to a competing center whose patients is ranked #2.

 

But 11 OPOs around the country are served by only one center. This gives that center huge flexibility to determine who gets what kidney, because if they refuse an offer for whoever is at the top of their list, they can still give the kidney to the second person on their list, or third, or 30th, theoretically.

 

But in practice, does this phenomenon, known colloquially as "list diving," actually happen? This manuscript from Sumit Mohan and colleagues suggests that it does, and at rates that are, frankly, eye-popping.

The Columbia team used data from the Scientific Registry of Transplant Recipients to conduct the analysis. The database tracks all aspects of the transplant process, from listing to ranking to, eventually, the transplant itself. With that data, they could determine how often, across these 11 OPOs, the #1 person on the list did not get the available kidney.

The answer? Out of 4668 transplants conducted from 2015 to 2019, the transplant centers skipped their highest-ranked person 3169 times — 68% of the time.

This graph shows the distribution of where on the list these kidneys went. You can see some centers diving down 100 or 200 places.

 

Transplant centers have lists of different lengths, so this graph shows you how far down on the percentage scale the centers dived. You can see centers skipping right to the bottom of their list in some cases.

 

Now, I should make it clear that transplant centers do have legitimate discretion here. Transplant centers may pass up a less-than-perfect kidney for their #1 spot, knowing that that individual will get more offers soon, in favor of someone further down the list who will not see an offer for a while. It's gaming the system a bit, but not, you know, for evil. And the data support this. Top-ranked people who got skipped had received a lower-quality kidney offer than those who did not get skipped. But I will also note that those who were skipped were less likely to be White, less likely to be Hispanic, and more likely to be male. That should raise your eyebrows.

Interestingly, this practice may not be limited to those cases where the OPO has only one transplant center. Conducting the same analysis across all 231 kidney transplant centers in the US, the authors found that the top candidate was skipped 76% of the time.

So, what's going on here? I'm sure that some of this list-skipping is for legitimate medical reasons. And it should be pointed out that recipients have a right to refuse an offer as well — and might be more picky if they know they are at the top of the list. But patient preference was listed as the reason for list diving in only about 14% of cases. The vast majority (65%) of reasons given were based on donor quality. The problem is that donor quality can be quite subjective. And remember, these organs were transplanted eventually so they couldn't have been that bad.

Putting the data together, though, I can't shake the sense that centers are using the list more for guidance than as a real mechanism to ensure an equitable allocation system. With all the flexibility that centers have to bypass individuals on the list, the list loses its meaning and its power.

I spoke to one transplant nephrologist who suggested that these data should prompt an investigation by the United Network for Organ Sharing, the body that governs all these OPOs. That may be a necessary step.

I hope there comes a day when this issue is moot, when growing kidneys in the lab — or regenerating one's own kidneys — is a possibility. But that day is not yet here and we must deal with the scarcity we have. In this world, we need the list to prevent abuse. But the list only works if the list is followed.

F. Perry Wilson, MD, MSCE, is an associate professor of medicine and director of Yale's Clinical and Translational Research Accelerator. His science communication work can be found in the Huffington Post, on NPR, and here on Medscape. He tweets @fperrywilson and his new book, How Medicine Works and When It Doesn't, is available now.

https://www.medscape.com/viewarticle/992716