Access and R&D, World Health Organization, Johns Hopkins University, Nature, Costa Rica, Germany, United States
development

Organisations. World Health Organization WHO, Johns Hopkins University JHU, Nature
Period. 14 – 20 May 2020

WHO to create a COVID-19 Intellectual Property Pool

The WHO embraced a proposal by Costa Rica to develop a voluntary mechanism that would pool patents and regulatory testing data to develop drugs, vaccines and diagnostics for COVID-19. Modelled after the Medicines Patent Pool by UNITAID for HIV/AIDS medical products, the COVID IP Pool would serve as a repository that target governments, industry, universities and non-profit organizations in order to ensure equitable access to resulting medical products, something that the Access to COVID-19 Tools (ACT) Accelerator announced last month does not sufficiently address.

The WHO has until May 29 to finalize the framework for the voluntary pool. The US has expressed reservations, while it has been accused of “pandemic chauvinism” because it refused to attend the launch of the WHO ACT Accelerator and is alleged to have offered to buy a German vaccine developer, while the CEO of Sanofi said that the U.S. may get priority access to COVID-19 vaccine, even though the company is headquartered in France.

Covid-19 has exposed the limits of the pharmaceutical market model

As the global search for Coronavirus treatments and vaccines intensify, two articles this week highlight the limitations of the existing drug development models. Health Policy Watch featured an interview with the Director General of a trade association representing the leading pharmaceutical companies. Among the “old-new tensions between profit and public needs” that were discussed was the “concern” by the pharmaceutical industry that the newly adopted World Health Assembly resolution referenced WTO flexibilities that allow countries to override patents to ensure access to vaccines, therapeutics and diagnostics.

The STAT article, meanwhile, points to failures of the existing market model that do not sufficiently incentivize investment into new treatments and vaccines for neglected diseases and pandemic preparedness despite warnings. Building on an extensive body of research, the author advocates for a new economic model of drug development that does not rely on market mechanisms. That would include more open and transparent research and development practices that address the “rapid global expansion of intellectual property” as well as more expansive government initiatives to address collective challenges (particularly given the significant role of government-funded research).

The 15 drugs being tested to treat COVID-19 and how they work

Even under the most optimistic projections, developing new treatments and vaccines for COVID-19 will require far greater time and resources than would be required for repurposed existing therapeutic compounds. An article in Nature this week provides an overview of the 15 leading candidates of existing drugs that are being studied as treatments for COVID-19. These can be divided into two categories: those that target the viral replication cycle, and those that aim to control the symptoms of the disease. They include:

  • Chloroquine/hydroxychloroquine (primarily used to treat malaria)
  • Lpinavir and ritonavir (HIV/AIDS)
  • Nafamostat and camostat (pancreatitis)
  • Famotidine (heartburn)
  • Umifenovir (influenza virus A and B)
  • Nitazoxanide (anti-infective)
  • Ivermectin (anti-parasitic)
  • Corticosteroids (cytokine storm, acute respiratory distress syndrome)
  • Tocilizumab and sarilumab (cytokine storm, ARDS)
  • Bevacizumab (oncology)
  • Fluvoxamine (antidepressant)

Contact tracing apps: balancing privacy and public health

Traditional public health contact tracing systems require significant time and resources, as one projection by Johns Hopkins University estimates a workforce of 100,000-300,000 would be needed to contain COVID-19 in the U.S. alone at a cost of $12 billion. As a result, health systems are increasingly turning to mobile contact tracing apps that are pitched as a way to reduce these costs by automating the process of logging people’s contacts should they became infected in the future.  Despite their promise, there are outstanding challenges, including marginalizing vulnerable populations that may not have smartphones.

There are also privacy concerns that has led civil liberties advocates to raise warnings about potential abuses, including by authoritarian countries during and after the pandemic. The CSIS also urges policy makers that digital tools such as contact tracing apps are not “silver bullet” substitutes for investment in fundamental pandemic response capabilities. That message was echoed by health authorities in Iceland, where Rakning-19 yielded very little benefit compared to traditional public health systems, despite its high uptake of 40% of the population.

Machine learning applications

In Nature Machine Intelligence, authors report on machine learning tools applying pattern classification to identify three biomarkers associated with COVID-19 mortalities.

Another paper in Nature examine real-time tracking of self-reported symptoms to predict potential cases of COVID-19. The smartphone app allowed the authors to develop a model that was 80% accurate and showed that loss of smell and taste were additional symptoms indicative of COVID-19. While this study used simple statistical regression analysis, another paper in pre-print used machine learning algorithms on a smaller Brazilian dataset that achieved better performance of 92% specificity and 95% sensitivity.

This development is part of the digest;