409 ‒ Inside modern drug development: the science, economics, and regulatory hurdles
In a Nutshell
Lloyd Klickstein explains that successful drug development requires identifying true unmet medical needs, then systematically minimizing dozens of scientific, regulatory, and commercial risks before Phase 3, because late-stage failures are catastrophic. Bimagrumab, developed to treat sarcopenia by blocking myostatin and activin receptors, increased muscle mass 4-8% in humans but only improved strength or function when combined with exercise or nutrition; the program was spun out and later acquired by Eli Lilly after showing additive fat-loss and lean-mass preservation with GLP-1 agonists. The same principles now guide new efforts to pharmacologically prevent epithelial cancers by activating the ribotoxic stress pathway that sorafenib inadvertently suppresses.
These notes were generated by AI and may contain inaccuracies.
Lloyd Klickstein is a physician-scientist who completed an MD-PhD degree. He trained in medicine and rheumatology, practiced rheumatology for approximately 10 years, and maintained an NIH-funded research laboratory at Brigham and Women's Hospital conducting basic science research on adhesion molecules in the immune system.
After his entire academic career at Brigham and Women's Hospital, he transitioned to industry over 20 years ago, joining Novartis Institutes. This move was initiated when Mark Fishman was recruited by Novartis CEO Dan Vasella to reimagine research and early clinical development in industry, bringing the concept of translational medicine to industry for the first time.
A common misconception exists about how new drugs are discovered. Steve Rosenberg's book "The Transformed Cell" addresses this by noting that intelligent people can have blind spots about drug discovery, questioning whether researchers simply experiment with random substances.
Klickstein starts with patients and clinical indications, looking for drugs or therapies that don't exist but are needed. There are two broad categories of drug development:
- Incremental improvements: Making drugs that are taken less frequently, converting injectables to oral formulations, or improving existing drugs (examples include atorvastatin and rosuvastatin in the lovastatin/simvastatin world)
- Quantum steps: Creating therapies for indications that may not have been described yet, lack ICD-10 or ICD-11 codes, and require establishing new regulatory pathways
The biggest value to society comes from creating new therapies rather than incremental improvements, though this involves significantly more uncertainty and failure.
Mark Fishman tasked a team with identifying the most needed medicines and determining what the company wasn't doing that it should be doing. The team received a budget to pursue this initiative.
They created a list of approximately 7,000 clinical indications that were unmet. This list served as a framework for thinking, despite containing only recognized indications at the time. The indications were grouped into:
- Things already being worked on
- Rare genetic developmental conditions difficult to approach
- Seven or eight different buckets including healthy aging, ENT, renal diseases, liver diseases, and fibrotic diseases
The program eventually grew to dozens of projects before being scaled down. One project that emerged was bimagrumab (Bimagrumab) for muscle diseases, stemming from the medical need of frail elderly people.
As a rheumatologist, Klickstein observed that patients who had to go to nursing homes faced approximately 90% three-year mortality rates. Being a frail elderly person requiring nursing home care was worse than most cancers in terms of clinical outcomes. There are many serious diseases in medicine that have worse outcomes than cancer, but they don't receive equivalent attention or treatment approaches.
Small molecules represent industry jargon for chemicals. Historically, major drug companies originated as dye companies over 100 years ago, as the chemistry for making dyes is similar to making drugs.
Biologicals encompass everything that is not a chemical, including:
- Antibodies
- Other proteins like peptides or soluble receptors
- Gene therapies (now considered a separate category with varying complexity based on delivery or targeting)
Devices are regulated separately by CDRH at the FDA and include:
- Workshop-made gadgets
- Apps
- Simple items like syringes or auto-injectors
- Implantable devices
Making a widely-used drug takes longer, costs more, and requires as many people as building the Burj Khalifa. Unlike buildings that can pay for themselves over decades, patent law provides limited time to recover investment and generate profit. At patent expiration, the drug becomes freely available for anyone to manufacture indefinitely.
Patents provide a 20-year term from filing, with possible extensions based on development time. Practically, this yields 10-15 years of exclusivity from launch. Patent law is the same regardless of drug type, but regulatory exclusivity periods differ between small molecules and biologics, changing periodically.
Some drugs have been protected by trade secrets rather than patents:
- Armour Thyroid: Desiccated thyroid hormone purified from pig thyroids before synthetic production was possible. The process was kept secret, making it technically difficult for competitors to replicate the exact composition including peptides and impurities needed for FDA approval using the same clinical package.
- Acthar Gel: Purified from pig pituitaries, containing ACTH. Used as a "rheumatology smart missile" for acute gout, providing an endogenous steroid taper over several days with a single injection. It was withdrawn during the BSE scare due to concerns about endogenous pig viruses, then acquired by a small company that re-commercialized it for infantile spasms at 100-1000x higher prices.
Companies may patent multiple aspects of successful drugs to extend protection. AbbVie with Humira patented composition of matter, formulation, salts, auto-injector, indication (method of use), dose, route of administration, combination with other drugs, and manufacturing methods. They staggered patent filings over time to effectively extend exclusivity.
Format selection considers both medical conditions and infrastructure. Big companies are format-agnostic, working with small molecules, biologics, gene therapies, and cell therapies. Medical rationales may favor specific formats:
- Childhood diseases require oral formulations that taste good
- Inhaled drugs suit specific lung conditions
A comprehensive drug development course would take approximately one year at Harvard seminar level. The process begins with identifying a medical indication and target, then determining drug format while minimizing risks at every step except those specifically identified as acceptable unknowns.
The worst outcomes in drug development are failing in phase 3 or succeeding in phase 3 but failing commercially. Failing early is critically important.
Sarcopenia is defined as decreased muscle mass with impaired muscle function, with ongoing debate about measurement methods (grip strength, gait speed, stair climb). The patient population consists of older adults, requiring suitable drug formats.
In drug development with new targets and indications, approximately 20 risks exist across target, format, bioavailability, toxicity, and other factors. Each risk at 90% success probability multiplies to near-zero overall success probability, necessitating risk minimization at every step.
Two strategic approaches exist for muscle-related drug development:
- Large indication approach: Targeting sarcopenia to help the maximum number of people, accepting potentially harder regulatory pathways since sarcopenia may not be clearly identified as a disease
- Orphan disease approach: Starting with Duchenne muscular dystrophy for faster regulatory approval, then expanding to larger indications once efficacy is demonstrated
Klickstein's personal bias favors large indications to help as many people as possible within limited time, though he has pursued both strategies depending on circumstances and drug mechanisms.
Initial sarcopenia drug development attempts focused on fall prevention. Falls have approximately 11 different causes including weakness, dizziness, vision problems, and attention issues. Dizziness itself is complex. Other factors include loss of reactivity, foot speed, proprioception, and vision.
The distinction between stumbling and falling relates to reactive speed of feet to catch oneself, involving type 2A muscle fibers. Multiple systems undergo atrophy creating conditions where minor challenges result in falls.
In clinical environments, only falls that cause injury are typically ascertained, as these are the patients physicians see. Patients are often reticent about reporting falls because they fear being removed from their home environment if deemed unsafe. A study was developed using triaxial accelerometers manufactured by a large company, creating a research device worn as a pendant. The study design included putting the device on bad ice skaters in Boston during winter, videotaping the rink as positive controls for actual falls, then examining device telemetry for sensitivity and specificity. The second phase planned to place the device on elderly nursing home residents, where the positive control was finding individuals on the floor who were unable to get up themselves.
The first part of the ice skating study was cut during institutional funding processes, leaving only the older adult component. Working with geriatrician Lou Lipitz and clinical operations person Kieran Dole, the study enrolled 60 subjects in one living environment, followed for six months. These participants had fallen at least once in the prior six months, making them high risk despite institutional protocols to prevent falls. The study captured 117 events where someone was found on the floor, but the device detected only 17% of real falls while generating 17% false actuations, rendering it not useful for measuring falls.
An attempt was made to use WiFi-type devices from MIT professor Dina Katabi to measure movements in homes, but this approach could not be implemented. The program for developing drugs to prevent falls was ultimately stopped because falls could not be objectively measured. Concerns about data quality included recall bias in frail older adults, whether participants would have writing materials available, and limited device familiarity at the time.
The discussion transitions to sarcopenia, muscle mass, and strength as the genesis of bimagrumab (BYM338) and activin receptor antagonists. Myostatin, discovered by Se-Jin Lee in the mid-1990s, is an inhibitor of muscle growth. Blocking myostatin in rodents produces dramatic muscle growth, as seen in knockout mice, chickens, dogs, and cows. In humans, the biology is more complex, involving both myostatin and activins, which is why inhibiting both or blocking the receptor provides therapeutic advantage.
Myostatin signaling occurs through the TGF-beta superfamily of receptors, including type 1, type 2, and type 3 receptors that signal via SMAD transcription factors. These govern gene programs affecting muscle size through regulation of muscle protein synthesis versus turnover. The proteins involved in muscle protein turnover include MuRF1 and MAFbx/atrogin. Myostatin signaling via activin receptors suppresses proteins involved in targeting muscle proteins for degradation.
Follistatin and follistatin-like proteins are endogenous inhibitors of the myostatin pathway. While gene therapy with follistatin produces larger muscles in rodents, it is a small protein with short half-life requiring multiple daily doses. The cost would be prohibitive, potentially exceeding a million dollars monthly. It was unclear whether follistatin would work in adults, as effects might require administration during a critical developmental window rather than in mature muscle tissue.
FC fusion proteins attach therapeutic proteins to the FC region of antibodies, extending half-life by hijacking endogenous recirculation mechanisms for blood proteins. The technology was originally developed by Brian Seed at Massachusetts General Hospital, with etanercept (Enbrel) as the first drug using this approach. Etanercept was initially tested in sepsis where it worsened outcomes, but proved highly effective in rheumatoid arthritis. The FC fusion extends protein half-life and enables easy purification using protein G and protein A columns. The original patents have expired, making this technology freely available.
Bimagrumab (originally BYM338) development began with Chris Lu's team and David Glass's team at Novartis Institutes for BioMedical Research, with Jeff Porter leading the pathways group. The strategy focused on inhibiting receptors rather than all possible ligands, recognizing that myostatin alone was insufficient in humans. Therapeutic antibodies were selected due to the nanomolar to low nanomolar affinity of ligands for receptors, which would be difficult to achieve with small molecules.
Antibody screening utilized phage display technology rather than mouse immunization, testing thousands of antibodies initially. A reporter assay using firefly luciferase was developed because activin type 2 receptors are expressed at levels too low to detect by flow cytometry. The assay measured decreased luminescence when myostatin or activin binding was blocked. The goal was to identify antibodies with higher affinity than the natural ligands, working on both myostatin and activin regardless of which ligand was present.
Identifying candidate antibodies took years, involving extensive biological work, tool development, and reagent preparation. Multiple candidates (typically 2-10) advance from screening, requiring confidence in developability including stability, predictable physical properties, and shelf life. Using fully human antibodies rather than humanized versions minimizes immunogenicity risk, though this remains one of the biggest unknowns in antibody development.
Fully human antibodies are made starting with human genetic material, but they still contain some foreign sequences. Antibodies have an intrinsic ability to recombine and create new sequences. In vivo in people, if a brand new antibody is not suitable, it gets selected out. In vitro work in a test tube does not have this selection process, so researchers use common codon usages and common antibody sequences to minimize issues, though the final behavior is only confirmed when tested in people.
In silico screens examine peptides likely to be generated in the proteome and assess whether they will have high affinity binding to MHC molecules. These screens are performed, but human testing is still required to confirm safety and efficacy.
The first antibodies tested in humans were mouse antibodies, and one is still used today as a therapeutic. OKT3 is an anti-T-cell antibody targeting CD3 that is used to prevent transplant rejection. It affects every T-cell. Rabbit anti-thymocyte globulin is also still used clinically. Antivenoms represent another category of foreign antibodies still used, most of which are derived from horse serum. A second exposure can trigger a ferocious serum sickness response.
Drug candidates must cross-react and work in at least one of the two species used for toxicology studies and in a species used for pharmacology. If the drug does not work in required species, surrogate drugs must be made. Bimagrumab worked in rodents. It proved immunogenic in mice but not in rats. The team created a murinized version called CDDH866 to enable mouse experiments.
Bimagrumab administration produced muscle hypertrophy in mice that exceeded the phenotype of pure myostatin knockouts. The mice were stronger and could run faster. The observed muscle mass increase reached approximately 30 percent. The team also examined whippet dogs and Belgian blue cattle as examples of natural double-muscled phenotypes.
Muscle function testing confirmed increased strength and running speed. Human results later showed only a 4 to 8 percent increase in muscle mass, with most tested individuals being older adults. David Glass performed experiments in older rats that showed efficacy, though less pronounced than in younger animals. Rodent muscle experiments are typically conducted in males because the response is stronger in males than females.
When developing drugs with abuse potential, companies work with WADA (World Anti-Doping Agency) early to ensure screening assays can be developed. The team initiated contact with WADA as soon as the antibody worked in mice. WADA has had an assay for more than ten years, though whether it has detected use remains unknown.
Therapeutic antibody programs often include non-human primates because one of the two toxicology species must demonstrate the expected pharmacology. This allows evaluation of both on-pathway and off-pathway toxicity. Preliminary studies in a small number of animals were conducted long enough to observe muscle hypertrophy, which occurred but was less pronounced than in rodents.
Toxicity assessment uses a weight-of-evidence approach based on all accumulating data. Clinical decision-making balances risk and benefit. A drug unsuitable for a broad population might still be appropriate for higher-medical-need patients, such as moving from sarcopenia to Duchenne muscular dystrophy. Two critical questions are whether toxicity is monitorable and whether it is reversible. Irreversible cardiac or neurologic toxicity typically ends development. Monitorable and reversible toxicity, such as liver enzyme elevations with terbinafine (Lamisil), can be acceptable if warning signs appear early enough to allow discontinuation.
Idiosyncratic liver toxicity remains the most common reason drugs are withdrawn from the market. It cannot be reliably predicted pre-clinically.
Large drug companies use two to four major checkpoints where data are assembled into slide decks, presented, and reviewed. Programs are adjusted based on feedback. Small companies make similar decisions but at higher frequency with faster cycles. At each stage, manufacturing scale-up, toxicology, and pharmacology studies proceed in parallel.
For a startup, raising capital for the pre-IND stage through phase 1 would require approximately 20 to 30 million dollars to reach phase 2a with runway for the next round. The value-creation milestone would be demonstrating muscle hypertrophy in the first clinical study. Large companies allocate fixed research budgets across hundreds of projects, while small companies concentrate capital on fewer programs.
DMPK (distribution, metabolism, pharmacokinetics) studies determine whether a medicine reaches its intended site and produces expected effects. Animals receive courses of therapy with blood tests, X-rays when warranted, and terminal autopsies to examine all organs for microscopic changes. These data must be obtained before first-in-human administration.
From first candidate to IND filing, attrition for therapeutic antibodies is relatively low at roughly five-to-one or six-to-one, with about 30 percent reaching human testing. Antibodies generally lack off-target adverse effects because of their high specificity; developers screen extensively to confirm lack of off-target binding. Small molecules can produce unpredictable off-target effects such as blood pressure elevation seen with early CETP inhibitors.
An IND (Investigational New Drug application) grants regulatory permission to administer an experimental drug to people. Patents are typically filed when a group of lead candidates has been identified, balancing the desire for maximum patent life with the need for protection once information becomes public. The IND application includes manufacturing data demonstrating GMP (Good Manufacturing Practice) capability.
GMP requires high-quality standards with extensive documentation proving identity, purity, activity, sterility, and absence of contamination. Multiple companies are typically involved: one manufactures the drug substance, another formulates it, a third packages it, and a fourth distributes it. The entire chain is inspected and documented. Online-sourced peptides lack these controls; buyers cannot verify contents or sterility, analogous to purchasing opioids from street dealers where fentanyl or other contaminants may be present.
BPC-157 exemplifies problematic research peptides: all published data originate from a single investigator, the sequence is not encoded in the human genome, and no receptor has been identified. The fundamental principle that real scientific findings are reproducible has not been satisfied. The premium paid for approved pharmaceuticals includes the cost of GMP manufacturing controls.
No off-pathway safety or tolerability issues were observed in animal studies for bimagrumab. The only finding of note was cardiac hypertrophy in rats; however, when heart size was normalized to the greatly increased body size caused by muscle hypertrophy, the ratio remained normal. Regulators accepted the argument that larger rats should have larger hearts. A recovery period in toxicology studies showed that both muscle and heart size decreased after drug withdrawal, supporting reversibility.
Toxicology studies are designed to achieve exposures in animals that are higher than what will ever be expected in humans. This creates a safety margin of exposure. Weekly dosing was used in animal studies despite the drug having a shorter half-life in animals. Both peak and trough drug concentrations are measured in these studies, with the goal of having both exceed human levels.
In general medicine therapeutic indications, toxic effects and therapeutic effects occur at different exposures. However, in oncology, toxic effects and therapeutic effects may occur at the same exposure or even lower levels. The medical need is so great that toxicity is accepted in these cases.
Patient selection for phase one studies depends on the drug type. For oncology drugs, testing occurs on the most recalcitrant cancer patients who have progressed through every therapeutic option and have stage four disease. This represents a Hail Mary approach where both tolerance and potential efficacy through dose escalation are being tested.
For non-oncology drugs, the preference is to test in the cleanest population possible where any measured effects can be attributed to the drug rather than underlying disease. However, healthy volunteers should not be exposed to risks if possible. The risk threshold used is the probability of being struck by lightning in the US in a year, which is approximately one in 100,000.
If the risk of adverse events exceeds one in 100,000, the study should move to a population that may potentially benefit from the therapy to justify the risk. This principle is not mandated by the FDA but represents an industry standard. Over 30 years of observation, serious incidents in healthy volunteers occur approximately once every 10 years in drug studies.
The TGN1412 incident involved a therapeutic antibody directed against CD28, which is an activating receptor on T-cells. The antibody was an agonist that caused cross-linking of CD28 receptors, leading to extremely strong T-cell activation and acute cytokine release syndrome in healthy volunteers. Some participants died. This incident led to the implementation of sentinel patient dosing in first-in-human studies, where cohorts typically include six people receiving active drug.
Another notable incident involved a small molecule from BIA (Bial), though serious adverse events in healthy volunteers remain extremely uncommon overall.
Medical students have historically volunteered for studies, including arterial line placement, brain electrical activity mapping studies, inhaling radioactive microspheres, and lymphocytapheresis procedures. These experiences provided exposure to various research methodologies and procedures.
For bimagrumab, older healthy volunteers were selected rather than young adults. This population allowed measurement of drug effects on muscle mass and strength without confounding factors from diagnosable disease or concomitant medications.
Phase one assessments included standard blood tests plus specific measurements for bimagrumab: muscle mass via DEXA and MRI, plus soluble muscle proteins in blood including CK, aldolase, and LDH.
Three adverse effects were observed that are considered on-target: muscle spasms or cramps, acne (more common in younger populations), and GI symptoms including diarrhea. The GI symptoms tend to be first-dose related and less common with subsequent dosing.
The approach in first-in-human studies is to dose as high as possible while never exceeding exposures tested in animals. For antibodies, the highest feasible dose is often limited by technical factors including volume of administration, mass of drug, and measurable contaminants from cell culture production that must also stay within animal toxicology exposure limits.
Bimagrumab was administered once monthly initially, with dosing frequency determined by emerging pharmacokinetic data. The highest dose reached was approximately 50-100 milligrams per kilogram. After the single-dose study, three doses were administered in the first-in-human study.
Multiple doses are typically needed to observe pharmacodynamic effects. Phase one studies focus on safety and tolerability, with the transition to phase two pivoting toward efficacy assessment as the primary goal.
To complete phase one requirements, a report is submitted to regulators. Communication with regulatory authorities occurs throughout the process whether dealing with the FDA or overseas regulators.
Every country has different regulatory requirements. The three biggest challenges in clinical studies are recruitment, recruitment, and recruitment. Key factors in site selection include access to qualified investigators, supportive regulatory environment, and study costs. Popular countries for studies include Germany, the US, Australia, New Zealand, and increasingly China.
China has become more favorable for clinical studies with a regulatory environment that allows investigator-initiated studies with less supporting data than required for typical IND applications. This enables faster testing with access to large patient populations.
Taiwan offers excellent investigators, English-speaking staff, centralized clinical environments with many patients at limited sites, and regulatory standards similar to the US.
Australia has a clinical trial notification process rather than approval process, where safety is assessed by ethics committees and drug quality by regulators. The favorable exchange rate also makes studies cost-effective.
Studies conducted in one country can support applications in others. Data from Australian studies can be included in US IND applications, and vice versa. However, FDA oversight requires studies to be conducted in the US.
Phase one studies are typically conducted in one to three countries, with data then used to support phase two studies in multiple countries and eventual global development.
Japan requires phase one studies in Japanese people before larger studies can be conducted there. These are called ethnic sensitivity studies, with formal regulatory definitions of Japanese ethnicity. The rationale includes genetic background differences and average body size variations affecting dosing and exposure.
Ethnic sensitivity studies for Japan can be conducted in Hawaii or California with Japanese populations, or in Japan itself. China also requires ethnic sensitivity studies due to documented toxicity differences in Han Chinese ethnicity.
Novartis conducted approximately 16 phase two studies across different indications for bimagrumab. The drug reliably increases muscle size but does not significantly improve performance assessments. In rodents, mass increases of 20-30% were observed along with performance improvements, but human mass increases are limited to 4-8%.
There are trends toward greater muscle mass increases in males versus females and younger versus older subjects, though considerable variability exists. A nutrition study examined protein and calorie intake at three levels: half the recommended daily amount, the recommended amount, and one and a half times the recommended amount.
Higher protein intake correlated with greater muscle building within the tested boundaries. The recommended daily amount tested was 1.2 grams per kilogram. When subjects received only half the recommended daily protein and calories, they lost muscle mass, but bimagrumab prevented this loss.
During early development, efforts were made to co-develop a nutritional supplement component with Novartis's nutrition division. However, when Novartis sold their nutrition unit to Nestlé, this collaboration ended. This represents a missed opportunity to optimize the therapeutic approach through combined drug and nutritional intervention.
The myostatin inhibitor program showed muscle mass increases but no corresponding strength gains without resistance training. This outcome aligned with previous findings from IGF-1 agonists and androgen agonists, including SARMs, where muscle hypertrophy occurred without strength improvements absent exercise. A meta-analysis of the Novartis sarcopenia studies demonstrated a 9-meter increase in six-minute walk distance from a 4-8% muscle mass increase in older adults. Additional functional assessments included timed up-and-go tests and short physical performance batteries.
The final Novartis study examined bagramab in type 2 diabetics using 10 mg/kg monthly dosing for 48 weeks. This maximal dose produced expected muscle mass increases alongside substantial fat mass decreases and hemoglobin A1C reductions of 7-8% absolute. The mechanism appeared to involve both increased insulin sensitivity and enhanced glucose disposal capacity, achieved without requiring dietary or exercise modifications beyond standardized advice to reduce intake by 500 calories daily.
Novartis spun out the asset in 2021 after determining the effect size insufficient for their strategic priorities. The program was acquired by a new entity formed in February 2021, initially named Versuspio, which later became the company developing bagramab. The founding team raised $70 million against a desired $100 million target, with Atlas Venture and Medici participating in the syndicate. The initial development focus targeted older adults with low muscle mass, impaired muscle function, and obesity, with the hypothesis that preserving muscle during weight loss would be particularly beneficial.
The obesity drug development landscape was considered commercially unviable at the time, with previous obesity drugs having failed commercially. This changed dramatically with the November 2021 publication of semaglutide data, which demonstrated the first truly effective obesity therapeutic. The development strategy shifted to positioning bagramab as an adjunct to GLP-1 agonists rather than standalone therapy.
Mouse studies demonstrated additive efficacy when combining bagramab with semaglutide, tirzepatide, or liraglutide. The combination produced unprecedented results for both fat loss and lean mass preservation. This finding expanded the opportunity significantly, leading to recruitment of experienced leadership including CEO Mark Bzanski and CMO Ken Addi. The company organization shifted with the original founding CEO moving to president and CSO roles.
Study naming followed a B-themed convention: BELIEVE for phase 2, BECOME for phase 3, and BEHOLD for post-registration studies. The generic drug name bagramab derives from Beimma, an Indian god described as strong as 10,000 elephants, with the suffix "mab" indicating monoclonal antibody classification.
The BELIEVE study evolved from an original 24-week sarcopenic obesity study to a 72-week combination study with semaglutide due to the changing obesity landscape. The study employed a full factorial design with nine arms testing all combinations of low and high dose semaglutide, low and high dose bagramab, and placebo. This design was necessary because human combination effects were unknown and to assess potential adverse effects from the drug combination.
Since semaglutide was proprietary to Novo Nordisk and unavailable as drug substance, the commercial auto-injector presentation was used, making placebo creation impossible and requiring open-label semaglutide administration. Bagramab was administered intravenously for the fastest clinical entry, though an auto-injector was under development. The study enrolled 507 participants with the primary endpoint being body weight, though DEXA scans were performed on all patients and waist circumference was strongly considered as an alternative endpoint.
At 72 weeks, the high-dose combination group achieved 22-23% body weight loss, with fat loss reaching 45.7% of starting body fat - comparable to bariatric surgery outcomes. Functional testing included timed up-and-go, short physical performance battery, 30-second chair stand test, and grip strength, with grip strength selected for the main study due to its linkage to clinical outcomes. Strength improvements were small and variable.
An unexpected finding was an approximately 20% increase in LDL cholesterol, attributed to on-target effects on hepatic activin receptors affecting LDL clearance. No adverse glucose effects were observed, and glucose reductions occurred independent of semaglutide contributions in diabetic participants.
Eli Lilly acquired the company after the first tranche of Series B funding closed. Current development status includes ongoing studies listed on ClinicalTrials.gov, including a complex combination study with tirzepatide. One study was paused, though other trials continue.
Six-month off-drug follow-up data will be presented at EASD in September. Muscle mass gains from anabolic agents typically revert toward baseline upon therapy withdrawal, as demonstrated in rodent studies. The study included participants with metabolic syndrome to assess diabetic endpoints during withdrawal periods.
The proposed obesity treatment paradigm involves induction and maintenance phases, using combination injectable therapies to achieve categorical shifts from obese to non-obese status, followed by maintenance therapy potentially using lower-dose injectables or oral GLP-1 agonists. LDL effects could be monitored and treated with existing cardiovascular prevention strategies.
mTOR inhibition is considered likely geroprotective in humans due to highly conserved biology across evolution, though effect sizes are expected to be modest. The primary challenge remains achieving selective mTOR complex 1 inhibition, as rapalogs downregulate mTOR complex 2 with sustained exposure. Alternative approaches include combining catalytic and allosteric inhibitors.
Intermittent dosing strategies with agents like everolimus or sirolimus may achieve selective inhibition, though measuring tissue-specific effects in humans remains challenging without liver biopsies. Rodent studies show young animals downregulate mTOR with fasting while older animals do not, raising questions about intermittent fasting efficacy in elderly populations.
The industry is beginning to recognize the importance of real medical prevention rather than focusing solely on treatment. This represents a fundamental transition from a sick care system to a healthcare system. Implementation requires better primary care infrastructure and reimbursement codes for preventive visits. Current coding systems lack mechanisms to bill for cancer prevention services, creating institutional barriers to building preventive medicine practices.
New drug development emerges from literature analysis. Recent papers from the past 5-10 years document drugs that cause cancer, suggesting these compounds inhibit cancer-protective pathways. Most drugs function as inhibitors, making this pathway analysis particularly relevant.
Sorafenib, a multikinase inhibitor used to treat renal cell carcinoma and hepatocellular carcinoma, causes skin cancers in approximately 10% of older patients. These cancers mirror normal population prevalence patterns, primarily basal cell and squamous cell carcinomas. Pathway biology reveals that sorafenib inhibits the sensing kinase triggering ribotoxic stress response, a pathway that induces cell death. This pathway, when irreversibly activated, serves as the target for potent toxins including diphtheria toxin, sarcin, and ricin.
The innovation involves controlled activation of the ribotoxic stress pathway. Since constitutive pathway activity exists in people and its inhibition causes cancer, the hypothesis proposes that gentle pathway activation could prevent approximately 50% of cancers. Skin cancer serves as the initial target due to its prevalence matching all other cancers combined.
The target population includes adults aged 50 and older with at least five previous skin cancers, representing a 50% annual recurrence risk. A cohort of 100-120 participants enables testing of low-dose, high-dose, and placebo arms to measure cancer prevention outcomes in a phase 2 study.
While skin cancer prevention represents a significant medical need, the mechanism may extend to epithelial tumors. However, testing cancer prevention across multiple cancer types presents substantial challenges due to the rarity of cancer incidence events, making such studies nearly impossible before drug approval.
An alternative approach involves treating patients who have completed adjuvant therapy for stage three epithelial cancers such as colon or breast cancer. These patients show no evidence of disease but face 50% recurrence risk. Stratification can identify high-risk individuals for treatment studies.
Tool compound screening across 1,000 cancer cell lines at the Broad Institute revealed melanoma as the most sensitive tumor type. While mutational burden from UV exposure correlates with mechanism susceptibility, this correlation alone cannot fully explain the observed tumor sensitivity. Melanoma treatment represents a viable therapeutic application alongside prevention.
Intensive sunscreen use versus usual practice has demonstrated melanoma prevention in a large Australian study, proving that pharmacological prevention approaches are testable in large phase 3 studies.
Pharmacological prevention represents a new approach distinct from traditional carcinogen avoidance strategies. Lifestyle factors including alcohol limitation, smoking cessation, insulin sensitivity maintenance, and weight management in overweight individuals all contribute to cancer prevention. The Swedish Obesity Study provides observational evidence that successful obesity treatment through bariatric surgery prevents multiple cancers over decades of follow-up.
Beimma development continues at Eli Lilly with parallel testing alongside newer generation GLP-1 agonists. Multiple pharmaceutical companies are developing pathway inhibitors following presentation of believe data at national and international meetings.
Keep Peter Attia MD in your library
Save the videos and channels worth coming back to, and find them again in one place.
