Improving clinical trials with machine learning

Machine learning could improve our ability to determine whether a new drug works in the brain, potentially enabling researchers to detect drug effects that would be missed entirely by conventional statistical tests, finds a new UCL study published in Brain.

“Current statistical models are too simple. They fail to capture complex biological variations across people, discarding them as mere noise. We suspected this could partly explain why so many drug trials work in simple animals but fail in the complex brains of humans. If so, machine learning capable of modelling the human brain in its full complexity may uncover treatment effects that would otherwise be missed,” said the study’s lead author, Dr Parashkev Nachev (UCL Institute of Neurology).

To test the concept, the research team looked at large-scale data from patients with stroke, extracting the complex anatomical pattern of brain damage caused by the stroke in each patient, creating in the process the largest collection of anatomically registered images of stroke ever assembled. As an index of the impact of stroke, they used gaze direction, objectively measured from the eyes as seen on head CT scans upon hospital admission, and from MRI scans typically done 1-3 days later.

They then simulated a large-scale meta-analysis of a set of hypothetical drugs, to see if treatment effects of different magnitudes that would have been missed by conventional statistical analysis could be identified with machine learning. For example, given a drug treatment that shrinks a brain lesion by 70%, they tested for a significant effect using conventional (low-dimensional) statistical tests as well as by using high-dimensional machine learning methods.

The machine learning technique took into account the presence or absence of damage across the entire brain, treating the stroke as a complex “fingerprint”, described by a multitude of variables.

“Stroke trials tend to use relatively few, crude variables, such as the size of the lesion, ignoring whether the lesion is centred on a critical area or at the edge of it. Our algorithm learned the entire pattern of damage across the brain instead, employing thousands of variables at high anatomical resolution. By illuminating the complex relationship between anatomy and clinical outcome, it enabled us to detect therapeutic effects with far greater sensitivity than conventional techniques,” explained the study’s first author, Tianbo Xu (UCL Institute of Neurology).

The advantage of the machine learning approach was particularly strong when looking at interventions that reduce the volume of the lesion itself. With conventional low-dimensional models, the intervention would need to shrink the lesion by 78.4% of its volume for the effect to be detected in a trial more often than not, while the high-dimensional model would more than likely detect an effect when the lesion was shrunk by only 55%.

“Conventional statistical models will miss an effect even if the drug typically reduces the size of the lesion by half, or more, simply because the complexity of the brain’s functional anatomy–when left unaccounted for–introduces so much individual variability in measured clinical outcomes. Yet saving 50% of the affected brain area is meaningful even if it doesn’t have a clear impact on behaviour. There’s no such thing as redundant brain,” said Dr Nachev.

The researchers say their findings demonstrate that machine learning could be invaluable to medical science, especially when the system under study–such as the brain–is highly complex.

“The real value of machine learning lies not so much in automating things we find easy to do naturally, but formalising very complex decisions. Machine learning can combine the intuitive flexibility of a clinician with the formality of the statistics that drive evidence-based medicine. Models that pull together 1000s of variables can still be rigorous and mathematically sound. We can now capture the complex relationship between anatomy and outcome with high precision,” said Dr Nachev.

“We hope that researchers and clinicians begin using our methods the next time they need to run a clinical trial,” said co-author Professor Geraint Rees (Dean, UCL Faculty of Life Sciences).

Clinical Trials of Antibiotics on Children

Central Licensing Authority i.e. Drugs Controller General (India) has given approval for conduct of various clinical trials of Antibiotics on children under one year of age. During the last three years, such clinical trials approved were mainly related to trials in Multi Drug Resistant Tuberculosis (MDRTB) and Tuberculosis Meningitis in children. The details of the clinical trials are registered in Clinical Trial Registry of India (CTRI), which is publicly available (www.ctri.nic.in).

Antibiotic use is a major driver of resistance. Neonates are more prone to infections and vulnerable to ineffective treatment. Sepsis remains a leading cause of mortality and morbidity, especially during the first five days of life and in low and middle-income countries (LMIC).

Antibiotics are included in Schedule H and H1 to the Drugs & Cosmetics Rules, 1945, and, therefore, cannot be sold in retail except on and in accordance with the prescription of a Registered Medical Practitioner.

Indian Council of Medical Research (ICMR) has launched a programme on Antimicrobial Stewardship, Prevention of Infection and Control (ASPIC) in 2012. Functional infection control programmes not only cut down the rates of nosocomial infections, but also reduce the volume of antibiotic consumption and are a critical part of any comprehensive strategy to contain antimicrobial resistance (AMR). Further, a red line campaign has been launched to regulate over the counter sale of Schedule H antibiotics. The campaign is aimed at discouraging unnecessary prescription and over-the-counter sale of antibiotics causing drug resistance for several critical diseases including TB, malaria, urinary tract infection and even HIV.

The Ministry of Health & Family Welfare has also launched a programme named ‘National Programme on Containment of Antimicrobial Resistance’ to address the problem of growing AMR.

The Minister of State (Health and Family Welfare), Sh Faggan Singh Kulaste stated this in a written reply in the Rajya Sabha here today.

Diet rich in tomatoes cuts skin cancer in half in mice

Daily tomato consumption appeared to cut the development of skin cancer tumors by half in a mouse study at The Ohio State University.

The new study of how nutritional interventions can alter the risk for skin cancers appeared online in the journal Scientific Reports.

It found that male mice fed a diet of 10 percent tomato powder daily for 35 weeks, then exposed to ultraviolet light, experienced, on average, a 50 percent decrease in skin cancer tumors compared to mice that ate no dehydrated tomato.

The theory behind the relationship between tomatoes and cancer is that dietary carotenoids, the pigmenting compounds that give tomatoes their color, may protect skin against UV light damage, said Jessica Cooperstone, co-author of the study and a research scientist in the Department of Food Science and Technology in the College of Food, Agricultural, and Environmental Sciences at Ohio State.

There were no significant differences in tumor number for the female mice in the study. Previous research has shown that male mice develop tumors earlier after UV exposure and that their tumors are more numerous, larger and more aggressive.

“This study showed us that we do need to consider sex when exploring different preventive strategies,” said the study’s senior author, Tatiana Oberyszyn, a professor of pathology and member of Ohio State’s Comprehensive Cancer Center.

“What works in men may not always work equally well in women and vice versa.”

Previous human clinical trials suggest that eating tomato paste over time can dampen sunburns, perhaps thanks to carotenoids from the plants that are deposited in the skin of humans after eating, and may be able to protect against UV light damage, Cooperstone said.

“Lycopene, the primary carotenoid in tomatoes, has been shown to be the most effective antioxidant of these pigments,” she said.

“However, when comparing lycopene administered from a whole food (tomato) or a synthesized supplement, tomatoes appear more effective in preventing redness after UV exposure, suggesting other compounds in tomatoes may also be at play.”

In the new study, the Ohio State researchers found that only male mice fed dehydrated red tomatoes had reductions in tumor growth. Those fed diets with tangerine tomatoes, which have been shown to be higher in bioavailable lycopene in previous research, had fewer tumors than the control group, but the difference was not statistically significant.

Cooperstone is currently researching tomato compounds other than lycopene that may impart health benefits.

Non-melanoma skin cancers are the most common of all cancers, with more new cases — 5.4 million in 2012 — each year than breast, prostate, lung and colon cancers combined, according to the American Cancer Society.

Despite a low mortality rate, these cancers are costly, disfiguring, and their rates are increasing, according to the U.S. Department of Health and Human Services.

“Alternative methods for systemic protection, possibly through nutritional interventions to modulate risk for skin-related diseases, could provide a significant benefit,” Cooperstone said.

“Foods are not drugs, but they can possibly, over the lifetime of consumption, alter the development of certain diseases,” she said.