The Year in Review / The Year Ahead

Dec 17, 2019 | News and Trends

“The only constant is change,” attributed to Heraclitus ca 500 BC, holds true for medical publication professionals in the 21st century. At times, it may feel like every week brings an announcement of a new regulation, guideline, protocol, or requirement, and we must always adapt, adjust, and move forward in kind.

“The Year in Review/The Year Ahead” spotlights notable topics in 2019 and considers the potential impact for the coming year(s). Topics were suggested by The MAP Working Group, then matched with subject matter experts. Commentary and perspective clarify each topic, offer insight into the practical implications for medical publication and medical communication professionals, and can serve as a guide to help us to assess, adapt, inform, and progress. This year’s topics are:

  • Clinical Trial and Plain Language Summaries
  • Artificial Intelligence in Medical Communications
  • A Heightened Focus on Open Access

The opinions expressed within are the authors’ own and do not necessarily reflect the views of the authors’ employer(s).

Clinical Trial and Plain Language Summaries

Jan Seal-Roberts, Adis, Springer Healthcare, London, UK; Sarah Griffiths, PhD, Oxford PharmaGenesis Ltd, Oxford, UK

Interviewed by: Jason Gardner, PhD, ISMPP CMPP™, CMC Connect, McCann Health Medical Communications, Macclesfield, UK

The primary purpose of regulatory clinical trial summaries (CTS) for laypersons and plain language summaries (PLS) within the peer-reviewed literature is to make science more accessible to interested parties. There are clear benefits of using scientific knowledge to help empower patients and carers, as well as healthcare professionals (HCPs), with the information to guide and understand medical treatment decisions. Therefore, after much debate in recent years, we are glad to see an encouraging surge in the production of CTS and PLS. Yet challenges remain.

First, who are they for?

We need to be clear what we’re talking about and who the audiences are. In many sectors, incorrect use of terminology has led to confusion. CTS, as described in the European Medicines Agency (EMA) guidelines [1], are not only for clinical trial participants. EU regulations require study sponsors to provide a summary of results “in a format understandable to laypersons” to “help improve familiarity and comprehension by the general public, participants, patients, and others” [2]. For PLS, confusion is more prevalent, e.g. when they are described incorrectly as “patient lay summaries.” Actually, PLS are intended for everyone: from information-seeking patients, carers, and journalists, to time-poor HCPs. However, while there is clear guidance around scientific exchange to HCPs [3], regulations around PLS as a form of scientific exchange beyond HCPs require definition.

Second, how are they written?

For CTS, clear guidance on format and structure is provided by the EMA [1] and others [4]. However, there are some challenges in writing these (e.g. in presenting complicated safety data or describing efficacy results); generally, only the primary outcome(s) – or maybe key patient-reported secondary outcomes – are reported. This means that CTS provide a limited view of the data. Conversely, PLS effectively serve as a publication’s executive summary, and so have more latitude to include the headline data, context, and interpretation presented in the associated publication.

In 2019, a useful PLS Toolkit was published [5], which provided templates and a checklist. Looking ahead, more guidance is required for authors as the quality and appropriateness of PLS remain highly variable. Many authors outside the pharmaceutical industry may not have the benefit of medical writing support, and further education would be useful on how to communicate science succinctly to busy HCPs and lay audiences. Furthermore, authors may be conflicted about whether to produce a PLS, when many journals are not actively encouraging them (yet!). In 2020, we hope that PLS adoption will continue as its value becomes increasingly apparent – eventually making it commonplace to submit a PLS with every manuscript.

Third, where are they?

The current inconsistency in the discoverability of CTS and PLS is likely to be a barrier to more widespread adoption. For CTS, an EMA portal and database are still in development (expected 2020 [6]). In the meantime, many pharmaceutical companies are posting their layperson CTS on third-party websites (e.g. [7]) or on their own company websites. By contrast, PLS may be difficult to find [8] due to the absence of a single searchable database. However, inclusion within PubMed is now possible, and discussions are ongoing to identify other effective solutions. It’s corny, but maybe those misquoting Kevin Costner et al. have a point: “If you build it, they will come!”

As we move into 2020, we are hoping that increased adoption and improved accessibility of PLS and CTS will demonstrate their value – but metrics will be critical to gauge their reach, engagement, and impact. We look forward to seeing how this story unfolds.

References

  1. https://ec.europa.eu/health/sites/health/files/files/eudralex/vol-10/2017_01_26_summaries_of_ct_results_for_laypersons.pdf (accessed 28 Nov 2019).
  2. https://ec.europa.eu/health/sites/health/files/files/eudralex/vol-1/reg_2014_536/reg_2014_536_en.pdf (accessed 28 Nov 2019).
  3. FDA Memorandum – Public Health Interests and First Amendment Considerations, January 2017. https://www.regulations.gov/document?D=FDA-2016-N-1149-0040 (accessed 28 Nov 2019).
  4. Barnes, A. & Patrick, S. Pharm Med (2019) 33: 261. https://doi.org/10.1007/s40290-019-00285-0.
  5. https://www.envisionthepatient.com/plstoolkit/ (accessed 28 Nov 2019).
  6. https://www.ema.europa.eu/en/human-regulatory/marketing-authorisation/clinical-data-publication (accessed 28 Nov 2019).
  7. https://www.trialsummaries.com (accessed 28 Nov 2019).
  8. FitzGibbon H, King K, Magee R, Piano C, Wilk C and Gaskarth M. Where are biomedical research plain-language summaries (PLS)? Poster presented at ISMPP 2019 European Meeting. Available at: https://ismpp.memberclicks.net/assets/docs/Education/EuropeanMeeting/2019EM/Posters/20Poster.pdf (accessed 28 Nov 2019).

Artificial Intelligence in Medical Communications

Dheepa Chari, MS, Global Scientific Publications, Pfizer Oncology; Ken Truman, MA, MedThink Inc
Interviewed by: Richa Chhabra, MS, CMPP™, Novartis, and Sally Hassan, PhD, CMPP™, Parexel

On behalf of ISMPP, Richa Chhabra and Sally Hassan interviewed Dheepa Chari and Ken Truman about the opportunities that artificial intelligence offers in publications and the broader medical communications space. The following conversation has been edited for length and clarity.

How do you anticipate artificial intelligence (AI) to be used in the development of publications on a day-to-day basis, and what will its impact be on the overall scientific communication plan?

Within the next 3-5 years, AI can enhance the following key areas in medical publications:

  1. Gap analyses during the insight generation phase/scientific communications plan: To conduct a gap analysis, the medical writer must do a literature review, which involves the extraction of a large corpus (i.e. a collection) of data/articles. This process is time-consuming as the writer needs to read potentially thousands of extracted documents and their abstracts/titles to distinguish the relevant ones. The document classification feature of AI can be utilized at this stage, wherein the medical writer can classify a subset of these documents (e.g., 200-300) by labelling them manually and then train the machine to classify the remaining set of documents automatically. The classification model would then be applied to the remaining documents in the corpus.
  2. Ongoing literature surveillance: AI can serve as a useful tool to share the latest research data of any relevant topic of interest into one’s email inbox, so people are aware in real time of any strategic opportunities.
  3. Publication development: AI can help reduce the time required in the development of a systematic literature review or any review article by using unsupervised machine-learning techniques to categorize and summarize the published literature. The methodology that would be used is still in its early stages.

During the early stage of involving AI in any publication process, both the methodology and the deliverables need to be validated against the traditional approach.

Are any AI tools currently available for use by medical writers and how will the role of a medical writer evolve with the increased use of AI?

Currently, AI tools are being explored to understand how they can be used as a validation piece. AI is a hot field that has led to the proliferation of vendors offering AI solutions for a specific task, but diligence is required when assessing these solutions. For structured data documents, natural language generation (NLG) can be applied to write templated documents. The first draft of review articles could potentially be generated using machine learning and NLG.

AI can be utilized to perform labor-intensive tasks that require a relatively low level of cognition, so that a medical writer would have more time to perform high cognition tasks, such as generating deeper insights from AI-sorted data. AI would, hence, lead to an improvement in the quality of these deliverables.

How will AI impact the need for greater patient engagement within medical communication activities?

There is a need to communicate scientific publications in plain language for a better understanding of data by patients or community practitioners. Machine learning or NLG can certainly be of help in translating highly scientific information to a patient-friendly format, with the oversight of a plain language writer. Another area of opportunity is to understand the type of information patients are seeking within a specific disease area. Topic modelling (an area of NLG) can be applied to filter through a large volume of conversations happening on the different patient advocacy websites and social media to understand the specific themes being discussed to inform patient insights. Ultimately, we are trying to serve patients who need to know what is happening within a disease landscape.

Will medical communication professionals need to be retrained to adopt AI, and should this training already be starting?

There is a requisite to embrace AI technology. Thus, capability development training is required for all publication professionals to understand AI better, its potential utilities, how to best use it, and how to develop more strategic insight capabilities from it. At the same time, it is also important for data scientists to understand what medical writers do to provide the best solution to a problem. Companies should be prepared to not only take up the technology, but also to adapt their workflow so that AI is a core capability of what they do. Training to embrace the technology should be starting now so that it can be adopted rapidly as the technology matures.

A Heightened Focus on Open Access

Jayme Trott, PharmD, MBA, ISMPP CMPP™, Head, Global Medical Affairs Operations, Janssen Global Services
Interviewed by: Eric Y. Wong, PhD, MBA, ISMPP CMPP™, Janssen Scientific Affairs, LLC

  1. Can you please tell us about some of the important initiatives around open data access that took form in 2019?

It is so appropriate that I would be responding to these questions during International Open Access Week (Oct. 21-27, 2019)! (To learn more, go to http://www.openaccessweek.org.)

Since the very informative January ISMPP U webinar on Open Access (OA) was offered, there have been several successful initiatives rolled out in 2019. Just this week in late October, Open Pharma published its Position Statement on Open Access (https://openpharma.blog/position-statement-on-open-access/). In May, after incorporating feedback from the public, cOAlition S published their updated Plan S, which requires that from 2021, scientific publications resulting from research funded by public grants must be published in Open Access formats. Other organizations like SPARC ‒ the Scholarly Publishing and Academic Resources Coalition (https://sparcopen.org/) ‒ and Right to Research (http://www.righttoresearch.org/) continue to advocate for open access to research, especially for students and young investigators.

Finally, Medical Publishing Insights and Practices (MPIP) rolled out three exciting resources this year:

  1. Discussions around improved data access have been ongoing for years, so what is it about these initiatives that make them outstanding?

Although academics have been clamoring for open access to research results for many years, both publishers and the pharmaceutical industry have been slower to adopt policies that require open access. With many public funders supporting Plan S, publishers will have to find ways to implement free, immediately available open access, or at least be more transparent regarding services for which they require payment. A few pharmaceutical companies have financially supported Open Pharma, but they have not been quick to endorse the position statement. This could change if publishers of high-tier journals begin to offer open access and don’t charge disproportionately higher fees for industry-sponsored publications.

MPIP’s tools are great educational and practical resources for pharmaceutical industry publication professionals to better understand and navigate the OA landscape.

  1. What changes to the practice of open access publishing do you foresee in the next 1-3 years?

I believe that industry will increasingly adopt OA policies and that publishers will find new business models that support OA while maintaining their revenue base. As an interesting example, MIT Press has recently secured a public grant that will allow them to pilot a new model in which they ask libraries to subsidize their open-access monograph publishing through a subscription model. Read more at: https://www.insidehighered.com/news/2019/10/22/exploring-subscriptions-support-open-access-monographs.

This article was prepared by the author in her personal capacity. The opinions expressed within are the author’s own and do not necessarily reflect the views of Janssen Global Services or Janssen Scientific Affairs, LLC.

The MAP Working Group hopes that this year-end review encourages all medical publication and medical communication professionals to celebrate the successes and accomplishments of 2019, and gain insight and perspective to support 2020 planning. We look forward to your continued readership, comments, and support of The MAP newsletter.

Happy Holidays!

The MAP Working Group
Eric Y. Wong, chair; Erica Chevalier-Larsen, Richa Chhabra, Jason Gardner, Anna Geraci, Sally Hassan, Paul A. Petruzzi, Lana Vegman

%d bloggers like this: