Huang Xianqin,Chen Chuhan,Dong Li

(School of Business Administration,Shenyang Pharmaceutical University,Shenyang 110016,China)

Abstract Objective To understand the real-world study (RWS) and its application in domestic and foreign pharmaceutical device regulatory decision-making,so as to promote its further application in the decision-making of medical devices supervision in China.Methods Literature analysis and comparative analysis were used to compare the real-world evidence (RWE) and randomized controlled trials to obtain the development status of medical devices in China and the United States.Results and Conclusion It is found that the application of RWS in the field of medical devices has many advantages,which can make up for the deficiencies of traditional clinical data.In addition,the RWE has a wide range of applications in the field of medical devices in China.In recent years,the RWE has received extensive attention in the medical field in the United States.The FDA has issued related guidelines,and its applications have gradually formed a complete system.Research on RWE will become the development trend of clinical evaluation of medical devices in the future.While attaching importance to the development of the RWE,China should learn from the specifications of relevant foreign organizations to improve its development in the field of health.

Keywords:real-world evidence;randomized controlled trial;medical device supervision

With the advancement of science and technology and the development of electronic health records,realworld data (RWD),and real-world evidence (RWE)have become new medical decision-making tools,which are increasingly concerned by researchers.In August 2017,the U.S.Food and Drug Administration(FDA) issued “Using Real-World Evidence to Support Medical Device Regulatory Decisions”,which provided guidelines for researchers to use RWE to support medical device registration.This article introduces the relevant definitions of RWD and RWE and compared the application of RWE in the field of medical device supervision in the United States and China,aiming to provide reference for the further improvement of China’s health system supervision and decision-making.

1 Real-world definition and background introduction

1.1 Related definitions

RWD refers to data related to patients’ daily health status or diagnosis,treatment and health care[1].

RWE refers to the clinical evidence regarding the usage and potential benefits or risks of a medical product derived from RWD analysis through appropriate and adequate interventional studies such as retrospective or prospective observational studies or practical clinical trials.Benefit-risk clinical evidence[2].

Real-world study (RWS) refers to the process of obtaining data related to the health status and/or diagnosis,treatment and health care of the subjects under the real environment for the present clinical problems through analysis[3].

1.2 Background introduction

In the past 50 years,randomized controlled trials(RCTs),especially explanatory RCTs,have been used in the medical field to provide information on the results of interventions in an “ideal” environment,but the results often cannot be extrapolated to clinical practice.With the increasing complexity of people’s health problems,traditional clinical data can no longer complete the review and approval of medical devices to promote the development of health services and human health.Moreover,the data obtained under clinical controlled conditions are relatively strict,which is different from the actual conditions of the patients.Therefore,it is not conducive to the effective evaluation of medical devices before marketing.On December 7,2016,the US Congress passed the“21st Century Cures Act” to speed up the approval of drugs and medical devices,encouraging FDA to conduct research and use RWE to support regulatory decisions on drugs and other medical products.In 2017,the United States formulated relevant policies and technical guidance for the evaluation and supervision of medical devices[4-6].Among them,the post-marketing medical product monitoring system represented by “sentinel initiative” was emulated by many countries[7].

The concept of RWS has been rapidly developed since it was introduced to China in 2010.Nowadays,it has been integrated into various fields of medical decision-making,and is an important source for evidence-based health science and hygiene decisionmaking[8].In April 2019,the National Medical Products Administration (NMPA),based on the current status,listed “Methodology Research on RWD for Clinical Evaluation of Medical Devices” as one of the nine key research projects in China’s drug regulation.On January 3,2020,the NMPA issued the “Guiding Principles for Drug R&D and Evaluation Supported by Real-World Evidence (Trial)” as the No.1 document,clarifying the value of RWE in all aspects of drug regulatory decision-making.Using RWE to support drug R&D and review was standardized at the national level.Therefore,the application of RWE in the field of health has received great attention in China.In this context,this article comprehensively analyzes related documents to provide reference for the further development of RWS.

2 Reasons for using RWE to support the supervision of medical devices in China

2.1 The traditional medical device supervision system faces significant challenges

Evidence-based medicine emphasizes that evidence is one of the elements that affect modern medical practice.The best evidence,clinical experience,and patient selection should be fully taken into consideration in any decision-making of diagnosis and treatment[9].In the past,RCTs,that is,traditional clinical trials,were used to obtain evidence to support regulatory decisions on drug devices.However,traditional clinical trials have some shortcomings,such as small sample size,insufficient representation of research subjects,short follow-up time of patients and high cost.Besides,the research environment is too ideal and it can only provide short-term results of the specific subjects in a strictly controlled clinical environment.The research results cannot be popularized,which is unable to provide all necessary information on the safety and effectiveness of the drug when the drug is marketed.Meanwhile,it is unable to provide sufficient evidence to support the real clinical practice,and further research is needed before clinical application.Lastly,the exclusion criteria and intervention measures are different from actual clinical medical effects,and it is difficult to observe rare adverse events.Therefore,the traditional clinical trials are difficult to meet the needs of the actual clinical environment and cannot answer all the questions in the decision-making of medical device supervision.

In addition,because the controlled environment required by RCTs is strict with long cycle,it is hard to achieve rapid entry of innovative medical products into clinical practice for patients who urgently need them.After marketing,spontaneous reports of adverse reactions are used as a monitoring method,which has limitations in discovering and verifying the safety of medical devices[10].

2.2 Advantages of applying RWE to the supervision of medical devices

In today’s society,the rapid development of digital medical care records,the collection and storage of a large number of health-related data,various statistical analysis methods and the application of databases have emerged quickly.The use of RWD to collect and analyze patient health data is an inevitable choice.

RWE research has the advantages of wide population coverage,strong representativeness of objects,low cost,less time-consuming,easy operability,and diversified application methods.Since the data and evidence come from the real world and can provide information about the results of intervention in the real environment,the research results are more applicable to clinical practice. The RWE research emphasizes the comprehensive use of multiple data,such as hospital electronic medical records,health and medical data,medical insurance data,etc.,which makes it possible to obtain longterm clinical results data.Therefore,RWE generated by RWD analysis is expected to be used to prove the safety and effectiveness of medical devices,thereby accelerating the approval of medical devices,expanding the indications of drugs,and supplementing the evidence obtained from RCTs.Under certain circumstances,it can provide a basis for the decisionmaking of medical device supervision.

In addition,RWE is widely used in clinical evaluation of medical devices,such as to support the clinical evaluation of medical devices in the full life cycle and product registration.As a supplement to existing evidence,it can provide clinical evidence and data in the clinical evaluation of the same variety to construct a single set of target values.At the same time,it also supports the modifications of indications and scope of application,such as long-term safety and effectiveness evaluation of high-risk medical devices(such as implants),and clinical evaluation before and after marketing.

3 Application of RWS on the supervision of medical devices in China

3.1 Drug supervision

3.1.1 Providing evidence of effectiveness and safety for the registration and marketing of new drugs

RWS can provide evidence of drug safety and effectiveness by collecting data on the effects of drugs in clinical use.When there is no experimental group or it is difficult to obtain the corresponding parallel control group,RWE can also be used as an “external control” for single-arm clinical trials.For example,crizotinib capsules for the treatment of non-small cell lung cancer (NSCLC) are based on the results of single-arm trials.A total of 3 studies were compared before its approval for import registration in China.

3.1.2 Providing evidence for changes of the labels of marketed drugs

Although RCT can expand the indications and application range of drugs on the market,the research design of RCT is strict,the selected subjects cannot fully represent the actual population of patients.Besides,RWE can obtain the effects of drugs through electronic health records and hospital databases,and clarify the clinical advantages of the drug and the target population,thus providing evidence for the modification of the indications,applicable population,and administration schedule of drugs.

3.1.3 Providing evidence for post-marketing requirements or re-evaluation of drugs

Due to the short RCT research time and the small number of samples,it is facing great challenges to popularize the results in clinical practical applications,so further monitoring and evaluation of post-marketing drugs are still needed.After the drug is marketed,it is faced with a real medical environment.Through RWE,we can understand the long-term effects and adverse reactions of the drug.Besides,the effectiveness,safety and economy of the drug can be comprehensively evaluated,and decision-making adjustments can be made accordingly.

3.1.4 Providing clinical evidence for the summary of famous TCM prescriptions and preparations of TCM

Due to the imperfect evaluation system,some Chinese medicines with good curative effects cannot be approved by NMPA and they can only be used in hospitals.While RWS has become a hot spot in the field of drug supervision,it also provides new methods for clinical evaluation of Chinese medicine.For example,RCT and RWE can be combined to make retrospective or prospective observational research to replace phase I or phase II clinical trials in the development of Chinese medicine.And then,RCT or PCT is designed to further confirm the effectiveness of Chinese medicine so that we can have better understanding of the value of TCM[11].

3.2 Medical device supervision

3.2.1 For clinical evaluation of the same variety of medical devices

Since the R&D of devices is generally improved and perfected on the basis of existing devices,it is important to collect RWDs of the same types of medical devices on the market for the design,R&D and clinical application of declared devices.Medical devices of the same type can be used to verify the effectiveness and safety of the declared devices by collecting existing RWDs,which also helps to reduce unnecessary clinical trials[12].

3.2.2 For the formulation of clinical trial scheme

Before carrying out clinical trials,enterprises often need to conduct a lot of investigations and analysis to identify the risks and benefits in clinical practice,so as to screen out indicators of the effectiveness and safety of the trial scheme as well as the appropriate follow-up time.Standardized and well-designed database data can also be used as a control group to provide a supplement to the one-arm test.In addition,due to certain ethical issues RCT is not suitable to be carried out in certain groups of people,RWD can be used to infer the possible risks when used in special groups and guide enterprises to continuously improve risk prevention and control measures.

3.2.3 For medical device marketing approval and post-marketing monitoring

The use time of some implantable devices in patients is much longer than the follow-up time of pre-market clinical trials,and the long-term safety and effectiveness cannot be evaluated.These devices should be monitored continuously or post-certification research should be conducted after they are sold on the market[2].For example,a cardiac pacemaker needs to be used by patients for decades,and the followup is only half a year before the market.However,the mechanical failure occurs only a few years after implantation.The long-term safety and effectiveness remain to be studied.RWD is particularly important for implantable devices after the market.

3.2.4 For changes and improvements of listed medical devices

After medical devices are marketed,it is often necessary to consider their values in medical field,which requires the help of clinical data generated in the real world.The practice of clinicians may expand the application scope of medical devices,including the use of over-applicable people,overindications and over-environment.In clinical practice,experienced medical staff can explore and summarize existing diagnosis and treatment methods according to their experience and skills.Meanwhile,they can also feedback the advantages and disadvantages of the devices in use and the information on adverse events.The information on these over-indications and related clinical data provide important clues for supplementing new indications.

3.2.5 For the review of newly developed medical devices

RWD can be extracted as a control group for the review of newly developed medical devices only if the data is relevant and reliable.When conducting clinical research on newly-developed medical devices,relevant RWDs can be extracted from the information systems of similar products as a control group for comparison.When the RWD provided by the control group is proved to be relevant and reliable,it can give evidence support for the review of newly developed medical devices.

4 The application of RWS on medical devices supervision in the United States

In the past 30 years,with the establishment and perfection of data files such as electronic medical/health records,the use of RWD has increased several times.European and American researchers have done a lot of research on the RWE,especially in the United States.Since the United States promulgated the “21st Century Treatment Act” in 2016,RWE has been incorporated into medical products and regulatory procedures.The United States has successively released related documents and guidelines for RWE.So far,the United States has launched the “Real-World Evidence Program Framework” to plan a roadmap for real-world applications[13].

4.1 FDA has issued relevant guidelines to form a real-world system to support the supervision of medical devices

The FDA issued the “Using Real-World Evidence to Support Medical Device Regulatory Decisions– Guidance for Industry and Food and Drug Administration Personnel” on August 31,2017.They are the guidance documents on pre-market and postmarket data collection,risk-benefit assessment,patient information,and accelerating the approval of medical devices that meet medical needs,thus simplifying the process of bringing new technologies to the market and maintaining the reasonable evaluation of safety and effectiveness of medical devices.In addition,the FDA also issued and implemented the “National Evaluation System for Health Technology (NEST)”,and began to use RWD to identify safety issues more quickly and better understand clinical benefit-risk characteristics of devices.The FDA believes that the proper use of NEST can also help reduce the time and cost required to support the marketing authorization of FDA-regulated products,and meet the needs of postmarket research and reporting[14].

4.2 The regulatory background for using RWE in the United States

4.2.1 Comprehensive conditions for using RWE

When the quality of RWD used to form the RWE exceeds the quality threshold,the FDA will use RWE to support the regulatory decisions for medical devices.Sufficient quality threshold should determine the clear purpose of regulatory use.For example,specific registration information can be used for post-marketing supervision,but the specific registration information must ensure that the safety and effectiveness or substantial equivalence of medical devices and medical products can be guaranteed before marketing.

It can be seen that when using RWD to make clear regulatory decisions,it is necessary to consider the applicable conditions of RWD and some purposes of using it,which included the following items:(1) Forming hypotheses to be tested in prospective clinical studies;(2) As a priori history control in Bayesian trial or a data source in the hierarchical model or generating mixed data;(3) As a combined control group or a mechanism to collect data related to clinical research,thus supporting device approval in registration data or some other system data;(4)As evidence for identifying,proving or supporting the clinical effectiveness of biomarkers;(5) As being approved or obtaining a humanitarian device exemption,pre-market authorization application(PMA) or innovation application evidence;(6)Supporting the application for reclassification of medical devices according to Section 513 (e) or (f) 3 of the “FD&C Act”;(7) As evidence of expanding device label information to include additional indications or updating label information on safety and effectiveness;(8) For ongoing public health surveillance,the signals identified each time suggest that the medical device may have safety issues.Then,RWE can be used to study these signals to understand the reasonable corrective measures;(9) Post-research stipulated for approval or may exclude post-marketing regulatory requirements specified in Section 522 of the “FD&C Act”;(10) Under certain circumstances,to generate summary report of medical device and provide some post-market data to replace pre-market data.

4.2.2 The collection of RWD should meet the requirements of the investigation device exemption(IDE) in Title 21 of the federal regulations

The FDA stipulates that devices approved by IDE do not need to meet other requirements of the“FD&C Act” for medical devices on the market.The purpose of doing this is to “encourage,discover and develop effective devices suitable for humans that meet the requirements of protecting public health and safety and ethical standards”.In many cases,IDE approval is required before starting a clinical study.

IDE regulations are applicable to all clinical studies that can determine the safety and effectiveness of medical devices,and some exceptions are specified.The FDA allows RWD to be collected during daily patient care to provide relevant information on the actual use of the devices.This includes RWD collected when medical devices are used for an unapproved intended purposes.If the collected RWD does not affect the way the device is used,which exists in the normal medical care process,IDE may not be required.Since the FDA does not regulate medical practice activities,which may include the use of medical devices for unapproved purposes (only when the devices are used in a legitimate doctor-patient relationship in accordance with the requirements of a healthcare practitioner).However,if data is being aggregated to determine the safety and effectiveness of the device,and then this process may affect treatment decisions,then such treatment may not belong to normal medical practice activities,and IDE might be required at this time.

5 Comparative analysis of RWE research between China and the United States

China’s RWS lags behind in theory and basic principles compared with the developed countries such as the United States.On the basis of theoretical research,western countries have promoted the application and development of RWE in scientific decision-making,and worked out RWE standard research methods.The United States promotes the application of RWE in medical device decisionmaking and supervision by strengthening legislation,formulating guidelines and carrying out stakeholder cooperation.Compared with the United States,China does not have many examples of RWE application and medical device supervision.Therefore,we are expecting a sound research system.

The use of RWE in the United States is popular than in China.The United States was the first country to apply RWE to medical and health technology evaluation.In 2007,the FDA launched the “sentinel initiative”,issuing the “FDA Amendment”,and establishing a post-marketing risk identification and analysis database.In 2008,the FDA officially started the “sentinel initiative” for piloting the establishment of a nationwide medical product safety monitoring electronic system[15].In February 2016,the sentinel initiative was fully completed.In July 2016,the FDA issued the industry guidance on “Using Real-World Evidence to Support Medical Device Regulatory Decisions”.The “21st Century Cures Act” approved by the U.S.Congress on December 7,2016 clearly raised the concept and application of RWE to a legal level.As of April 2017,the number of real-world studies registered on the US clinical trails.gov website was 46 557,including 2 780 patient registration studies[16].

6 Outlook and recommendations

RWS can prove the long-term safety and effectiveness of drugs and medical devices,and it will become the focus of clinical evaluation of medical devices in the future.Before drugs and medical devices go on the market,RWE can be used as a useful supplement to clinical data,and at the same time it can provide evidence support for the development and improvement of the devices.After drugs and the devices go on the market,RWE can evaluate the actual effect of rare drugs,or unexpected serious adverse reactions,safety and effectiveness,etc.[17].Besides,RWE also has a broad prospect in pharmacoeconomic evaluation,such as pre-market pricing and negotiation for post-market medical insurance access.

Although RWS provides important information and powerful means in decision-making of drugs and medical devices,due to the difference of decisionmaking environment,the complexity of problems and the diversity of methods,RWE has certain challenges in the specific use.It is recommended that China build electronic health records and electronic medical records databases under the premise of ethics.The United States and other developed countries have many experience in RWS.Therefore,China should refer to the research results,technical specifications and guidelines of organizations such as FDA and ISPOR to set up a real-world application system suitable for China’s national conditions.

RWS also has limitations,because its data sources are extensive,resulting in uneven data quality.Besides,the research results are biased due to the baseline inequality between groups.It is recommended to develop a corresponding collection method and standard analysis system.In addition,since RCT is the best standard for measuring the effectiveness of treatment in the past,RWE and RCT can be combined together so that they can complement each other.Then,they can be applied to supervision and health technology evaluation to promote the development of drugs and medical devices,and even the entire health field.