The clinical studies landscape is evolving quickly: are you?
Innovation in clinical trials
Innovation in clinical trials
In the last decades, innovation has accelerated in the pharmaceutical sector, and new technologies are now ready to revolutionize clinical trials. Big pharmaceutical players are already paving the way for collaborations with startups and implementing innovations in their clinical trials.
Pharmaceutical companies of all sizes that are willing to jump onto the innovation bandwagon ought to draw a roadmap strategy, starting with four mature enablers: real-world data (RWD) analytics, the use of social media for efficient patient recruiting, digital health & point of care testing for accurate and timely diagnosis closer to the patient, and digital companions & ePRO. Once these enablers are implemented, pharmaceutical players can anticipate mid-term revolutions by using an External Control Armand genomics-based clinical trials. Finally, two long-term evolutions are still under development but could shape the future of clinical trials: digital twins and organs on chips.
To apply this strategic roadmap, identifying the right startups to partner with is key to building innovation capabilities. Execution of the partnerships must be taken seriously, as pharmaceutical players have long be enused to developing internal capabilities and collaborating with startups might not be part of their DNA.
Beyond the buzz words related to new trends in clinical trials (“In Silico”, “computational modeling”, “virtual trials”, ...), the clinical studies landscape is experiencing a radical paradigm shift. Indeed, the digitalization of the pharmaceutical industry is impacting clinical trials, slowly becoming 100% site-agnostic. Synthetic control arms and tests conducted with organ-on-chips even foreshadow a future for clinical trials and drug approval exempted from human testing.
The revolution is already occurring, strongly impacting clinical trials and threatening pharmaceutical companies who won’t adapt quickly to fall behind. This white paper deciphers the trends and technologies that are making the future of clinical trials.
Trends show that R&D costs of developing a new drug roughly double every nine years, 70% of which are dedicated to clinical development, and 30% to pre-clinical drug discovery. Moreover, clinical success rates are constantly decreasing, from 22% in the 1980’s to a little more than 9.6% in 2018. These trends are proving the Eroom’s law (Moore’slaw spelled backwards) right. Observed since the 1980s, it states that drug discovery is becoming slower and more expensive over time, despite improvements in technology (such ashigh-throughput screening, biotechnology, combinatorial chemistry, and computational drug design). Therefore, the overall productivity of drug development is constantly decreasing, which has a strong impact on pharmaceutical players.
Meanwhile, preclinical productivity is improving, driven by multiple enablers and innovations. For instance, Artificial Intelligence (AI) technology, that relies on key advances in biology and computation, allows to identify more targets and drug candidates, in a shorter time frame: what used to take 4 to 5 years a while ago takes less than a year today. Likewise, the average compound time to go from synthesis to initial human testing has dropped from 4.3 years in 2000 to 2.6 years in 2020. Although these key evolutions open opportunities for pharmaceutical companies, they also require highly efficient clinical development programs.
Today, the position of patients has shifted from mere “subjects” who generate data to informed collaborators. Pharmaceutical companies reckon that their participation is key to the overall success of clinical trials. This realization has led to the emergence of the concept of “patient-centric trials”. Indeed, technology, social media and self-education play a crucial role for patients today, in relation to their own healthcare.
Pharmaceutical companies have started involving patients from the initial stages of clinical development. This early implication helps them to build protocols and study designs that are more real and closer to life experiences. Pharmaceutical companies are also investing in enhancing the focus on patient education, engagement and retention, designing educative websites and using social media platforms. As a result, these “patient-centric trials” are proving to be more cost-effective.
The innovation revolution in clinical trials is happening now because some technologies have reached their maturity level. Healthcare companies can rely on technological bricks such as cloud computing and AI to analyze huge sets of data to gather decisive insights to set or adapt their clinical trials. This effect is increased by the fact that on the other side, healthcare institutions are raising their standards in terms of digitization and are making extensive data available. This trend will continue, as regulators push for more standardization in terms of reporting and of data availability. (see below : RWD).
Additionally, the miniaturization of devices is making home testing possible, with high accuracy and a large set of parameters that can be tested with only a single drop of blood for instance. Technological bricks such as natural language processing (NLP) are starting to be used in the conception of digital companions used in clinical trials and in the analysis of social media in the participant screening and selection phase.
Regulation can be a brake to change, but in the case of clinical trials, the United States Food and Drug Administration is leading the way for regulators. Recently, the FDA has been promoting innovative clinical development programs. Since 2018, the FDA has launched pilots to advance innovative clinical trial designs as part of the agency’s broader program to modernize drug development and promote innovation in drugs.
Indeed, “Modernizing clinical trials is an agency-wide priority” stated, Scott Gottlieb, M.D. and the Former FDA commissioner, in 2019. “Without a more agile clinical research enterprise capable of testing more therapies or combinations of therapies against an expanding array of targets more efficiently and at lower total cost, important therapeutic opportunities may be delayed or discarded […] More trials can incorporate data from electronic health records, and adopt electronic informed consent, to enroll more patients in clinical trials closer to where they live and work.” By promoting innovation in clinical trials, the FDA intends to enable faster access to therapies.
COVID-19 has strongly impacted clinical trials, increased remote interactions and triggered new FDA guidelines.
Since last year, the Coronavirus pandemic has stopped or reoriented most clinical trials. 55% of trials active pre-COVID are currently paused or cancelled, and in April 2020, 30% of new clinical trials were concerning COVID-19. In addition, in March 2020, the number of patients entering trials had dropped by 65% compared to the previous year. Pharmaceutical companies have had to think about alternatives to cope with the situation.
The unprecedent situation has engaged a new dynamic in the care industry. Remote interactions have increased 4 times with patients.
In the meantime, the FDA has issued a “22 Questions and Answers to address Covid-19 Crisis for Clinical Trials” document. It shares guidelines and pieces of advice on administrative matters concerning clinical trials, offers the opportunity to modify the clinical trial protocol and promotes the use of new practices to ensure clinical trials continuity. Among them, the FDA officially enables remote trial, to support clinical trial operations with online and remote processes. It also allows the continuation of the clinical study in case of limitations of travel due to external factors(COVID-19) or patient personal conditions.
International pharmaceutical companies have taken the plunge and started collaborating with startups. Indeed, innovative startups and technological new entrants can save years of R&D for incumbents, which have the finances, network and capabilities to scale projects.
These collaborations are already taking place in the fields of data query to pre-identify patients, off-site study technologies, or even the use of predictive analytics to develop new therapies.
For instance, Sanofi is collaborating with TriNetX to reduce clinical trial complexity and speed up development, via a cloud research platform using “electronic health records (EHRs), including demographics, diagnoses, procedures, medications, labs and genomics, to reduce trial design complexity, catalyze recruitment, and help streamline trial investigator work”.
Likewise, the startup Nerve Liveis helping Novartis in drug development processes, analyzing huge amounts of data and using this knowledge to research and develop new therapies. This partnership is expected to also contribute to a broader agenda related to the way information is structured, stored and shared across the company, with the objective of unlocking the power of data that was previously stored in silos.
As disruptive technologies are becoming more mature, both patient and process-wise, pharmaceutical companies must embrace this paradigm shift and change their practices in terms of clinical trials. Starting with the most mature innovations, these trends will shape the future of clinical trials for years to come.
In thefollowing matrix, we grouped the 8 major elements of innovations in clinicaltrials into 3 separate blocs.
Let’s dive into these 3 blocks ofinnovations:
Innovation in clinical trials is notably driven by 4 major enablers:
These key innovations must be initiated early on to build up the foundationsof a cutting-edge clinical trials strategy.
Real-World Data (RWD) collects and structures data that is generated outsideof the controlled environment of a clinical trial. It includes Electronic Health Records (EHRs), insurance claims data, disease registries, historical clinical trials, data generated from at-home patient monitoring devices, and other sources that can be relevant for the clinical trial.
RWD is a powerful way to draw a holistic view of a patient from themolecular level to the overall health state. This holistic picture of the patient further enables personalized medicine. Moreover, analyzing Real-World Data helps gather patient-level insights and build evidence to support regulatory-grade decisions about treatment safety, efficacy, and cost-effectiveness. RWD can therefore be of great help in clinical trials.
Aetion Evidence Platform: delivering real-world evidence for life sciences, payers, providers, and regulatory agencies
Aetion platform enables manufacturers to transform large and disparate data sets from real-world evidence into conclusions needed to bring treatments to market faster. Moreover, manufacturers can partner with payers to enter value-based contracts built on shared and trusted evidence. Aetion clients can increase their productivity while maintaining high levels of scientific quality and transparency.
Aetion partnered with Pfizer on different topics (for precision oncology to renal cell carcinoma, prostate, breast, and lung cancer), as well as with Johnson & Johnson Janssen division on a multi-diseases AI project (characterization of individual cancers).
Aetion also collaborates with the FDA to validate applications of Real-World-Data and modernize regulatory guidelines.
Social media are increasingly becoming an important component of clinical trial patient recruitment strategies. Indeed, recruiting participantsfor research studies can be difficult and costly. Studies show that less than10% of patients in the United States enroll in clinical trials, and even lowerrates are noted among minority populations.Traditional recruitment strategies such as print media show poor results interms of reach (especially with hard-to-reach populations) and effectiveness ofthe campaign. In comparison, the popularity of social media platforms (e.g.Facebook, Instagram, Snapchat, …) and their usability, allow for a bettertargeting of patients.
In fact, social media offer the potential to use dynamic communicationmechanisms to reach large, diverse populations, therefore increasing awarenessand ultimately accrual to clinical trials. However, if patient recruitment onsocial media proves to be more advanced, user-friendly, and effective thantraditional media recruitment strategies, studies advise to use a combination of both to obtain representative samples.
Point-of-care testing (POCT, or bedside testing) is defined as a medical diagnostictesting at or near the patient point of care. It consists of runninga clinical grade screening or a diagnosis test outside of the laboratorysetting, whether it be the physician’s office, the pharmacy or at-home, by thepatient. The main benefit of POCT lies in its ability to bring an accurateand timely diagnosis closer to the patient, accelerating the “diagnosis-treat cycle”.
Initially focused on infectious diseases in developing countries, POCTuse cases are expanding to new areas. In oncology for instance,POCT enables pre-treatment testing at the point-of-care, allowing the physicianto determine the next treatment steps and regimen choices immediately withoutwaiting for further results. POCT also allows at-home monitoring oftreatments’ side effects and adverse events.
More complex POCT is being used specifically in clinical trials, withdevices allowing patient enrollment in a single visit while reducing failurerates. POCT devices can be directly connected to the centralized clinical trialsystems, allowing for enhanced communication and constant monitoring of thetracked parameters.
Athelas: delivering remote care programs
Athelas is an American biotech startup. The 50-person team has developed a low-cost Point of Care (POC) monitoring system that measures complete blood count (CBC) parameters from a single drop of blood in seconds. The device is indicated for use for quantitative determination of white blood cells (WBC) and neutrophil percentages (NEUT%), as well as lymphocytes, platelets, etc.
Athelas is especially useful for at-home rapid point-of-care testing or in the context of clinical trials. The device can be used for the identification and monitoring of chemotherapy-induced neutropenia (CIN) for patient care and clinical trials.
The Athelas One is now FDA cleared for point-of-care indications. An independent comparison study between Athelas device (finger prick) and the gold standard Sysmex 5000 (venous blood sample) showed 0% clinical range error of Athelas and its superiority in clinical classification.
Patients may face many challenges and a tortuous road from diagnosis to treatment and long-term follow-up. This can be the case for patients diagnosed with cancer, for instance. However, caregiver systems are understaffed and notable to offer continuous and personalized support to these patients. In this context, digital tools such as digital companions and Electronic Patient Reported Outcomes Software (ePRO) are perfectly suited to fill the gap: they increase patient support and improve patient's quality of life (QoL), especially during clinical trials.
Indeed, these digital tools can:
In particular, digital companions can play a key role in the patient journey. They can provide patients with information, through personalized content, digestible formats (like video, image, or text), and automated repetitive communication. They can also be of great help in managing treatments and tracking symptoms, sides effects and mood swings. Moreover, digital companions help personalize assistance for patients and provide peer support and community connections, that are crucial for the patient journey and quality of life.
Digital companions have the potential to significantly impact clinical trials. In the recruiting phase, patients are much more likely to participate in a study if they have the possibility to raise their concerns or ask questions at any moment. Digital companions can also help to collect even more data than conventional clinical trials, in real-time, and can improve cost efficiencies, to shorten timelines, as well as to increase patient satisfaction and retention.
Kaiku Health: optimizing care through timely symptom management and improved workflow
Kaiku health is an oncology digital health companion and platform, allowing personalized support and improved Quality of Life for patients. Kaiku’s engine enables intelligent symptom screening and reporting, care team alert, and personalized patient support. The platform allows the capture of real-world data strengthening the personalization of care and the transition to value-based care.
Founded in Finland in 2012, Kaiku’s platform is CE mark as Class IIa Medical Device. Kaiku Health is currently used in clinical routine by thousands of cancer patients for 25 different cancer types. The startup is also developing specific solutions to manage toxicities of Immune Checkpoint Inhibitor and CAR T Cell therapies.
Kaiku Health is collaborating with Roche to develop digital patient support for cancer immunotherapy, and with Amgen for digital symptom tracking in multiple myeloma.
Once these four enablers are implemented correctly, pharmaceutical companies should focus on the next two trends that are technology-ready: External Control Arm and Genetic Testing. These two innovations are to revolutionize clinical trials in the next 5 years.
Clinical trials historically imply a large number of patients to form a randomly allocated control arm, essential in proving the efficiency of the treatment or drug in trial. External Control Arms are trying to streamline clinical trials, by reducing the number of patients required, thus lowering costs and accelerating timelines.
Indeed, an external control arm is a data-rich way to use existing patient-level data from past clinical trials, combined with real-world data (RWD) sets, to augment the performance of a randomly allocated control arm. A combination of a well-constructed external control arm with a small, randomized one could significantly reduce the number of patients needed and accelerate bringing therapies to market.
The FDA strongly recommends implementing genomics into studies:
“With advances in science and increased awareness of the impact of genomics, there is a need and an opportunity to maximize the value of the collected samples and the data generated from them. Therefore, genomic sample acquisition is strongly encouraged in all phases and studies of clinical development.”
Indeed, genomics-based clinical trials can be used to improve trial success and accelerate the pace of clinical research. Incorporating genetic testing into clinical trials can help to shed light on the varying patient responses and identify sources of variability in drug response. According to the FDA, “the identification of genomic biomarkers underlying variability in drug response may be valuable to optimize patient therapy, design more efficient studies and inform drug labelling”. For instance, genetic differences between participants of a clinical trial can enable scientists to identify biomarkers associated with the clinical patient outcome. The accountability of sources of variability can play a key role indrug development.
Moreover, when recruiting patients for a clinical trial, utilizing genomics could help separate patients into sub groups. This patients tratification facilitates the identification of the best candidates for aspecific clinical trial, further speeding up trials and eliminating ineffective therapies at an earlier stage. Once again, this can help accelerate the pace and reduce the costs of market research. Patient stratification is likely to become a standard, as it improves treatment efficiency.
However, using genetics in clinical trials requires a lot of specific expertise, from data analytics, bioinformatics, and counsellors, and can increase the costs. Moreover, using genomic data in clinical trials isn’t always justified from a patient perspective (for instance for treatments that are not influenced by the patient’s genetics). Further training and increasing the understanding of the potential of genomics is key to decide when to invest in implementing genomics into trials.
If the technologies of Digital Twins and Organs on chips are not fully available yet, these two innovations encompass a high disruption potential for clinical trials.
A digital twin is a multi-scale “in silico” representation of a person. This virtual replica of an individual can be of great use for clinical trials. Indeed, digital twins ideally reﬂect the person’s “state of being” through high levels of granularity of metabolism, physiological, behavioral and morbidity information. As such, they could enable extremely precise personalized health protection and medical care.
In the field of drug development, a panel of thousands of drug compounds can be used to virtually treat digital twin copies and to predict individual risk-beneﬁt proﬁles. These predictions can help better design clinical trials, by identifying individuals and sub-groups expected that most beneﬁt from a speciﬁc drug treatment.
Thus, digital twins could greatly increase the accuracy and cost-eﬀectiveness of clinical development programs – with an important impact on both drug quality and costs.
The term “organ-on-chips” refers to a cell culture-based model system inwhich cells of different kinds are placed on small structures (chips). These chips are usually made of synthetic polymers that have been “patterned” into grooves, channels, or spirals, just like electrical chips.
The combinations of cells and chips can model the smallest functional subunit of an organ or tissue in the human body, in a miniaturized format outside the body. Organ-on-chips can mimic the alveoli of a lung, a small number of synchronously contracting heart cells, or even a mini kidney and liver-like structures. They can also model human disease states, by adding bacteria, immune cells, drugs, or even cells from diseased tissue to the chips.
This technology offers tremendous flexibility and robustness in drug screening and development through the improvement of predictivity of drug response in human tissues or organs. By providing the possibility to develop personalized tissue or organ models, organs-on-chips let us foresee a future ofclinical trials without patients.
As big pharmaceutical players are moving forward, the entire industry is called to innovate, adopt new technologies, and shake up the way clinical trials are conducted today – and those who won’t jump onto the innovation bandwagon will irretrievably fall behind.
The leap is big, yet by developing their capabilities to collaborate with startups, pharmaceutical companies of all size can speed up their innovation path. Collaboration may not be part of pharmaceutical players’ DNA,but with the right support from innovation specialists, they can:
Whatever we do, it is always in close collaboration with the business teams (clinical teams, scientific teams, etc.) who are the carriers of knowledge and needs. Our mission can be broken down into two pillars: the first is strategic; we must define the ambition and the digital transformation strategy with the teams, and then implement it. The idea is to set an ambitious medium / long-term goal that will significantly speed up our clinical studies. The second is more short-term and involves supporting studies which are kicking off today. Indeed, many solutions are now available in our service catalogue and depend on clinical protocols and study constraints.
Technology is an important element, but it must be coupled with the development of certain skills internally. I think that to anticipate a future transformation you need to:
This is a complex question to answer because it depends on many factors. I would have to say recent developments around synthetic control arms, as well as digital companions to collect more information from the patient, or digital biomarkers, are very promising.
We worked in parallel on three projects:
Our goal is not to become a "POC machine" (to avoid launching many unsuccessful projects) but to lay the foundations for a more fundamental transformation. We must be careful to properly define the contributions of advanced technologies such as IA, ML to ensure that they meet a real need.
In particular, we have developed several studies that highlight the relevance of a data-driven and patient-centred vision. These studies demonstrate that the exploration of the multitude of data generated from the discovery of a drug up to its use in real life after regulatory approval, allows us to produce edifying analyses of the epidemiology and clinical characteristics of neuroendocrine tumours (NETs). These include a retrospective study using data from five years of electronic health records, which aims to better understand NETs and to identify new therapeutic strategies [but also] the Phase III RAISE study, carried out in collaboration with Owkin, on the use of deep learning models and dissociated response to predict the effectiveness of early treatment in patients with NETs.
 DiMasi JA. Innovation in thepharmaceutical industry: New estimates of R&D costs. J Health Econ. 2016May;47:20-33.
Scannell,J. W., et al. Diagnosing the decline in pharmaceutical R&D efficiency.Nature Reviews Drug Discovery, 11(3), 191–200. 2012
Estimationof clinical trial success ratesand related parameters. Biostatistics. 2019 Apr; 20(2): 273–286.
 The Moore’s Law is a computing termwhich originated around 1970. It states that the number of transistors on amicrochip doubles every two years, though the cost of computers is halved.
 Zhavoronkov et al 2019,Stokes et al. 2020
 “Patient centric approach forclinical trials: Current trend and new opportunities” by Neha Shankar Sharma,July 2015.
 Statement by FDA CommissionerScott Gottlieb, M.D., on new strategies to modernize clinical trials to advanceprecision medicine, patient protections and more efficient product development.March 14, 2019
 McKinsey & Company –COVID-19 Crisis: Biopharma Response, Clinical Operations Roundtable, May 112020
 “Social Media and Clinical Trials:The Pros and Cons Gain Context When the Patient Is at the Center” by Michael A.Thompson, MD, PhD; and Ruth M. O’Regan, MD.
 “The Effectiveness Of Social Media(Facebook) Compared With More Traditional Advertising Methods for RecruitingEligible Participants To Health Research Studies: A Randomized, ControlledClinical Trial
TheEffectiveness Of Social Media (Facebook) Compared With More TraditionalAdvertising Methods for Recruiting Eligible Participants To Health ResearchStudies: A Randomized, Controlled Clinical Trial” by Mai Frandsen, Megan Thowand Stuart G Ferguson.
 E18 Genomic Sampling andManagement of Genomic Data Guidance for Industry, FDA, March 2018
 “Genomics in clinical trials: thebest current initiatives in chronic diseases”, Sano Genetics
 “Stem Cells - Scientific Facts andFiction”, second edition, 2014, by Christine Mummery, Anja Van De Stolpe,Bernard A.J. Roelen, and Hans Clevers.
 “Human-Derived Organ-on-a-Chip forPersonalized Drug Development”, Yasamin A. Jodat, Min Gyeong Kang, KiavashKiaee, Gyeong Jin Kim, Angel Flores Huidobro Martinez, Aliza Rosenkranz, HojaeBae, and Su Ryon Shin.