IPM Brief – Issue 18 | August Currents: The Science is Already in the Water

By Denis Horgan, Secretary General of the International Alliance for Personalised Medicine

August 6, 2026
Editorial

Good morning, and welcome to the August 6th edition of the IPM Brief.

For many readers – and particularly the lucky ones – August means holidays, quieter roads, slower inboxes and perhaps a swim before the day becomes too warm. Anyone entering the water this morning has an exceptional example to keep on the horizon.

A century ago today, Gertrude Ederle became the first woman to swim the English Channel. She completed the crossing in around fourteen and a half hours, beating the existing men’s record by almost two hours. August 6 is also the birthday of Alexander Fleming, born in Scotland in 1881, whose discovery of penicillin transformed modern medicine. (Reuters)

Ederle crossed a physical barrier. Fleming helped medicine cross a biological one.

Personalised medicine now faces its own channel: the distance between what science can reveal and what healthcare systems routinely deliver.

On one shore sit liquid biopsy, genomic testing, digital pathology, targeted therapies and increasingly sophisticated data. On the other sit patients whose access still depends on geography, reimbursement, infrastructure and whether a health system is ready to act.

Enjoy the quieter August rhythm where possible. But the science does not stop, and neither do the implementation gaps.


Europe speaks confidently about prevention. Its health systems remain organised, and largely paid, to wait for disease.

A new European study, drawing on 26 stakeholder interviews and 270 completed surveys, identified an interconnected set of obstacles to personalised prevention: health strategies focused on treatment rather than prevention; limited public and professional awareness; unresolved ethical, legal and social questions; and weak cost-benefit models. (Frontiers)

The barriers reinforce one another. An intervention that is not reimbursed is not routinely delivered. A service that is not delivered does not build workforce capability. And an approach without implementation data struggles to win investment, creating a circle in which promising prevention remains permanently “promising”.

The study is exploratory rather than population-representative. Participation was self-selected, respondents tended to be highly educated and responses were concentrated in Portugal, Italy and Sweden, although the survey included participants from 26 countries. But its central message is difficult to escape: Europe does not primarily have a prevention-innovation deficit. It has an implementation deficit. (Frontiers)

THE IPM TAKE: Prevention will remain the policy everyone endorses and too few patients receive until reimbursement, HTA, workforce planning, data governance and budgets reward disease avoided, not simply activity delivered.


Cancer medicine has a new problem: not too little data, but too much confidence in what the data appear to say.

Liquid-biopsy reports increasingly include variant allele fraction, or VAF. A patient with metastatic HR-positive/HER2-negative breast cancer might show an ESR1 mutation at 1% and a PIK3CA mutation at 10%. The temptation is to treat the larger figure as the more important target.

That would be a mistake.

Nicola Fusco and Umberto Malapelle argue that VAF is a contextual measurement, not a treatment-ranking system. It is shaped by tumour fraction, metastatic-site shedding, clonal architecture, copy-number changes, assay sensitivity, treatment pressure and the timing of the blood draw. (ScienceDirect)

The biology matters. PIK3CA mutations are generally earlier, clonal and stable. ESR1 mutations often emerge under endocrine-treatment pressure and can remain subclonal and dynamic. Their percentages are not directly comparable votes for therapeutic priority. The pivotal evidence supports the presence of an actionable mutation, not an arbitrary VAF hierarchy, as the basis for treatment selection. (ScienceDirect)

THE IPM TAKE: Laboratories should report VAF with biological context. Molecular tumour boards should resist unsupported numerical hierarchies. And payers should not turn unvalidated thresholds into access barriers. Measurement without interpretation is false precision.


DNA Double Helix Spiral Molecule Structure 3d illustration.

The longevity industry has spent years looking for a pill to slow ageing. Nature may be pointing further upstream: improve the machinery that protects the genome.

Bowhead whales can live for more than 200 years despite reaching more than 80,000 kilograms. Their cells show unusually efficient and accurate repair of DNA double-strand breaks, linked partly to high levels of the protein CIRBP. Increasing CIRBP expression in fruit flies extended lifespan and improved resistance to radiation. (Nature)

It is intriguing science, not a prescription for humans.

DNA repair is not a simple “more is better” switch. Some pathways protect healthy tissue; others can preserve damaged cells or help tumours withstand treatment. The real target would be the right repair, in the right cells, at the right time. The researchers also caution that much of the work used fibroblast models rather than the epithelial cells from which most human cancers arise. (Nature)

THE IPM TAKE: This is personalised medicine applied to ageing. Any future intervention will require biomarkers, long-term endpoints and cancer surveillance. The breakthrough is unlikely to be immortality. It may be preventing molecular damage from becoming decades of multimorbidity.


Washington wants a new scientific golden age powered by artificial intelligence.

Around 5,000 teams applied to the Genesis Mission, with 278 projects selected in its first phase. The programme aims to deploy AI across autonomous laboratories, drug discovery, energy, materials and infrastructure. The administration’s Science: A New Golden Age report also calls for faster funding, more ambitious national missions and greater use of AI throughout scientific research. (Nature)

There is much to welcome in that ambition. Science can be slow, fragmented and overly cautious. AI can improve experimentation, interrogate enormous datasets and accelerate discovery.

But there is a contradiction. Nature reports that hundreds of NIH applications that have already passed peer review are being subjected to additional political scrutiny. An algorithm checks proposals against 235 disfavoured terms, including “gender” and “climate change”, leaving some applications in administrative limbo. (Nature)

That is not merely science reform. It is portfolio control by vocabulary.

THE IPM TAKE: AI can find drug targets and interrogate genomes. It cannot, by itself, recruit representative populations, build trust around data or explain unequal uptake. A golden age cannot be built by accelerating politically approved science while filtering out inconvenient questions. Evidence, not a prohibited-word list, must determine what deserves investigation..


A genetic test for Parkinson’s disease can produce very different answers depending on whose DNA is being tested. Drug development has not caught up.

The largest multi-ancestry analysis of established Parkinson’s genes examined 99,783 people across 11 ancestry groups, including 58,559 people with Parkinson’s disease. Overall, 2.1% of patients carried an established causal variant, but the proportion ranged from 0.4% among participants of African ancestry to 10.7% among those of Ashkenazi Jewish ancestry. (PubMed)

The differences among risk variants were equally striking. GBA1 risk variants ranged from 4.1% among East Asian patients to 52.9% among African-ancestry patients, while LRRK2 risk variants were particularly frequent among East Asian patients. (PubMed)

These findings do not mean ancestry determines destiny. Risk variants were also found among controls. They mean Parkinson’s does not have one universal genetic map, and that the frequency and clinical meaning of variants can differ substantially across populations. Yet trials targeting GBA1 and LRRK2 pathways remain concentrated largely in Europe and the United States. (PubMed)

THE IPM TAKE: Diversity is not a recruitment box to tick after a therapy has been designed. It should determine which variants enter diagnostic panels, where trials open, how eligibility is defined and where genetic counselling is available. Trial geography must follow molecular need, not merely established infrastructure.


Latin America and the Caribbean face a substantial cancer burden among younger adults, but still lack a sufficiently complete map of who is affected, where and why.

An analysis across 32 countries estimated 269,881 cancer diagnoses and 85,695 deaths among people aged 15–50 in 2022. Early-onset disease represented 17% of new cancer cases and 11% of cancer deaths. Breast and cervical cancers dominated among younger women; testicular and colorectal cancers were the most frequently diagnosed among younger men. (AACR Journals)

Colorectal cancer provides the clearest warning. Mortality among younger men increased significantly in seven countries and among younger women in five. In Paraguay, the annual increase approached 4.5% for men and 5.9% for women. (AACR Journals)

This is not automatically an argument for screening every adult before 50. It is an argument for better intelligence and faster clinical recognition. Persistent symptoms should not be dismissed because a patient falls outside the traditional age profile. Countries need stronger registries, age-specific reporting and research connecting disease patterns with genetics, obesity, diet, environmental exposure and early-life risk.

THE IPM TAKE: Earlier diagnosis must connect to biomarker testing and appropriate treatment. The first rule of personalised medicine is to know whom the system is failing. Count younger patients properly, before they arrive too late.


The United States ended surprise medical bills. It did not end surprise medical costs.

The No Surprises Act has achieved its essential purpose: protecting patients who unknowingly receive out-of-network care. Evidence indicates that out-of-network claims and patient cost-sharing have fallen. But the cost has moved behind the curtain, into arbitration between insurers and providers. (Congressional Budget Office)

Regulators originally expected around 22,000 disputes annually. By May 31, 2026, more than 6.3 million had been initiated since the process opened in April 2022. Providers have prevailed in more than 80% of decided cases, while awards have often been far above conventional payment benchmarks. (CMS)

Doctors say insurers lowball initial payments and delay awards. Insurers say large provider groups and billing intermediaries have industrialised arbitration. The uncomfortable possibility is that both sides are exploiting weaknesses in a system designed around their restraint.

THE IPM TAKE: The patient shield must remain untouchable. But the machinery behind it needs credible benchmarks, eligibility checks, transparent decisions and enforceable deadlines. The bill disappeared at the bedside. Policymakers must stop the same cost returning through higher premiums.


Asthma is usually counted in inhalers, hospital visits and deaths. It should also be counted in missed education, lost work and reduced economic participation.

The number of people living with asthma is projected to rise from roughly 260 million in 2021 to around 275 million by 2050, largely because of population growth. Yet one of the most expensive failures is also the most basic: not delivering treatments that already work. (ScienceDirect)

A Global Asthma Network study covering 453,473 people in 25 countries found that, among those with severe symptoms, 44.8% of children, 60.1% of adolescents and 55.5% of adults were not using a corticosteroid-containing inhaler. Poor asthma control was markedly more common in lower-income settings. (PubMed)

Personalisation matters for patients whose asthma remains uncontrolled: phenotyping, biomarkers and biologics can transform care. But precision cannot compensate for absent basics.

THE IPM TAKE: Asthma policy needs two tracks. First, universal access to accurate diagnosis, affordable anti-inflammatory inhalers, adherence support and cleaner air. Second, personalised pathways for patients who remain uncontrolled. The dividend comes from matching the right intervention to the right patient, while ensuring everyone can reach the starting line.

That is enough heavy lifting for one August morning.

For those already on holiday, enjoy the slower pace. For those still at their desks, the weekend is at least visible from here. And for anyone heading for a swim, remember Ederle: the other shore may look distant, but distance is not the same as impossibility.

The science is already in the water. The task is to make sure it reaches the other side.


August is for holidays. September is for new ideas. 

While you enjoy a well-deserved break, whether by the sea or in the mountains, you can already plan what’s next. Registrations are now open for our events in New York (September 24), Stockholm (September 12) and Dublin (October 2).

Visit our Events page to explore what’s coming up and reserve your place.


Follow the last-mile conversation!

IPM Alliance tracks the decisions, bottlenecks and implementation gaps shaping personalised medicine access. Follow us for briefings, events, policy signals and practical insights from across regions and disease areas.


We want to hear from you

When a patient becomes eligible for personalised care, where does the system most often lose them: identification, testing, referral, reimbursement, delivery capacity or equity?


IPM Brief is built for people working where science, policy and patient access collide. If one colleague would find this useful, forward it to them and help bring them into the conversation.


International Alliance for Personalised Medicine

Avenue de l’Armée / Legerlaan 10, 1041 Brussels, Belgium

This email was sent to jayasinghtec29@gmail.com. You are receiving this email because you are part of the IPM Alliance network or have engaged with our activities, briefings, events, or policy work.