Characteristics of chief investigators and principal investigators in Australian and New Zealand radiation oncology clinical trials

Globally, the research community is coming to realise the need for diversity, equity and inclusion (DEI) amongst research teams and leadership. Diverse teams reduce homogeneous ‘group think’, propagate innovation, propound support for broader more representative research and facilitate the recruitment of patients from diverse backgrounds. Given the above, this study aims to retrospectively examine the characteristics of chief investigators (CI) and principal investigators (PI) in past and present Australian and New Zealand radiation oncology clinical trials.


Introduction
Clinical trials are prospective research studies whereby human participants are assigned to receive healthrelated interventions and then assessed for the impact of such interventions on their health outcomes. 1,2The execution of clinical trials typically involves a team of onsite support staff, health professionals, biostatisticians, study coordinators and data managers.Amongst these individuals, the chief investigator (CI) is the individual responsible for the overall development, conduction and reporting of a clinical trial whilst the principal investigator (PI) is responsible for carrying out the trial at a particular site as part of the research protocol. 3,4lobally, the research community is coming to recognise and value the need for diversity, equity and inclusion (DEI) within research teams and leadership.Diversity refers to the presence of differences within a given setting (e.g.race, ethnicity, sex, expertise). 5Equity refers to the fair provision of opportunities amongst individuals by strategically addressing any inherent imbalances between them. 5Last, inclusion involves creating an environment whereby individuals feel valued to fully participate in their roles. 5,6The benefits of DEI in team settings have been well established.8][9] In the context of clinical trial research, DEI amongst researchers may reduce homogeneous 'group think', confront standing assumptions and theories, and thereby lead to more accurate decision-making and improved problem-solving efficacy. 10,11Such outcomes, in turn, may challenge the status quo, inspire the next generation of research questions, diversify shared knowledge and practices and ultimately encourage innovation. 10,12otably, diversity is also important in perpetuating diversity.For instance, via the flow-on effect, diverse individuals in clinical leadership roles can enhance the recruitment of diverse trial investigators and teams whilst diverse representation within funding bodies and grant review panels can propound support and funding for broader and more representative research. 13t is important to note, however, that the success intrinsically associated with diverse teams should not be the sole driving force for DEI within clinical trial teams and leaders.Rather, it is also important to value an equitable system that recognises that talent is distributed across all groups and supports equal opportunities amongst individuals independent of their identity. 14iversity, equity and inclusion amongst clinical trial research teams and investigators may additionally lead to benefit patients.This is because diverse clinical teams are often better attuned to understanding and addressing the health outcome disparities amongst diverse patient groups. 12,15Furthermore, greater visibility of DEI within research teams also may promote trust, interest and acceptance amongst potential participants from diverse backgrounds. 10,12][18] Clinical trials have traditionally been carried out in large public academic health centres in metropolitan areas.However, there is recognition that extending trial availability to private and rural and remote centres is also important.This is because it facilitates equitable patient access to novel interventions, enables patients to receive more thorough monitoring and provides the opportunity for health services to gain site credibility. 19,20Last, given that many trials fall short of meeting scientific objectives due to low accrual rates, an added benefit to site expansion is that it often facilitates faster patient recruitment and thereby minimises trial costs. 19,21ancer is the leading cause of disease burden in Australia. 22According to the Australian New Zealand Clinical Trials Registry (ANZCTR), cancer-related clinical trials account for the 2nd highest number of clinical trials behind mental health-related trials in the past two decades. 23Radiation therapy is a cornerstone treatment modality for cancer with studies estimating that 52% of cancer patients would benefit from radiotherapy at least once during their course of illness. 24,25he Australian Government currently funds 14 multi-Site Collaborative Cancer Clinical Trials Groups to support Australia's capacity to develop independent, industry-led cancer clinical trials protocols.As one of these groups, the Trans Tasman Radiation Oncology Group (TROG) specifically functions to facilitate the development and conduct of radiation oncology-related research in Australia and New Zealand. 26This study aimed to examine the characteristics of the CI of TROG's clinical trials, as well as the characteristics of the site Principal investigators (PI) and associated institutions.It also aimed to identify potential opportunities to help improve DEI in future TROG clinical trials.

Procedures
This retrospective study was conducted to examine the characteristics of TROG clinical trial CIs from 1989 to 2022 and PIs from 2000 to 2022.Data on investigator characteristics were attained from trial protocols available on the TROG website as well as archived master database files provided by the TROG Scientific Committee.Data of interest included CI and PI discipline (e.g.radiation oncology, medical oncology, medical physics), clinical trial activation date, institution type (private vs. public) and geographical location if in Australia.For the latter, the Modified Monash Model (MMM) 2019 classification system was used to categorise Australian clinical trial institutions into metropolitan, regional, rural and remote areas.These areas, divided into seven categories by the MMM system, are based on town sizes and data from the Australian Bureau of Statistics (ABS). 27ublicly available data on the Australian Health Practitioner Regulatory Authority (AHPRA) website and the Medical Council of New Zealand (MCNZ) website was used to determine the registered self-reported sex of the CIs and PIs.For medical specialists, these websites were also used to determine the year in which they attained their medical fellowship.This year, alongside the year of clinical trial activation, was used to calculate a medical specialist's number of years of post-fellowship experience at the time in which they were appointed as a PI or CI.Years of experience were classified into the following categories: early career (<5 years), middle-career (5-10 years) and late career (>10 years).
If the above information was not available in the first instance, other institutional websites and personal networks were used to determine this data.Amongst clinical trials that had not been initiated at the time of this study, those without an activation date were excluded.

Analysis
CI and PI characteristics were collated and tabulated in Microsoft Excel.Descriptive statistics were employed to elucidate the data obtained.

Results
The characteristics of CIs and PIs who have led TROG clinical trials from 1989 to 2022 and 2000 to 2022, respectively, are detailed in Tables 1 and 2.
As can be seen in Table 1, 120 TROG clinical trials have been led by 134 CIs from 1989 to 2022.Of the 134 CIs, 86.6% (n = 116) were radiation oncologists (ROs), 9.0% (n = 12) were medical oncologists (MOs), 3.0% (n = 4) were medical physicists and less than 1% were radiologists and psychologists.At the time in which they were appointed as a CI, 19.1% (n = 25) had less than 5 years of experience post-fellowship, 30.5% (n = 40) had between 5 and 10 years of experience postfellowship and 50.3% (n = 66) had more than 10 years of experience post-fellowship.
Most trials have been initiated by CIs in the public (98.4%;n = 124) compared to private settings (1.6%; n = 2).All trials have been led by CIs from institutions located in metropolitan areas (100%, n = 126).
As can be seen in Table 2, 465 PIs have opened 463 TROG clinical trials within their respective institutions from 2000 to 2022.88.6% (n = 410) of trial sites were in Australia, 10.6% (n = 49) of trial sites were in New Zealand and 0.86% (n = 4) of trial sites were located internationally.Of the 465 PIs, 95.1% (n = 442) were ROs, 3.7% (n = 17) were MOs and less than 1% were haematologists, medical physicists and nuclear medicine physicians.At the time in which they were appointed as a CI, 19.4% (n = 25) were considered early career, 31% (n = 40) were considered middle career and 49.6% (n = 64) were late-career professionals.

Discussion
The value and need for DEI amongst clinical trial participants, research teams and leaders are garnering recognition amongst the research community.To date, no formal review had been reported on the diversity of TROG clinical trial investigators and their associated institutions.This study addresses this gap, and the findings could inform the strategic priorities of the organisation such that interventions to improve diversity in membership, investigator and leadership can be identified and implemented.Most TROG clinical trials have been led by RO CIs (86.6%) and PIs (95.1%).The predominance of RO investigators is unsurprising given their specialised role and expertise in prescribing and overseeing radiation therapy treatments for cancer patients, given that TROG clinical trials focus on radiation medicine-related research questions and given that TROG members would predominantly be ROs.The involvement of MOs as CIs (9.0%) and PIs (3.7%) is explicable given the complementing role of systemic oncology treatments for cancer patients.However, whilst medical physicists ensure the accurate delivery of all radiation therapy treatments, they only made up 3.0% and 0.4% of CIs and PIs, respectively, and notably, whilst radiation therapists are closely involved in the day-to-day care of patients undergoing radiation therapy treatment, none have ever led a TROG clinical trial.It is possible that this reflects the lower representation of these radiation medicine professionals in the organisation's membership overall.Future initiatives to support and mentor TROG members from non-medical disciplines to engage in research idea generation, concept development, protocol development and funding applications would be important.In addition, education, clarification and awareness of who can and who cannot be a site PI could be considered as it has been shown in other settings that the low involvement of non-medical health professionals as clinical trial investigators may be due to the unclear requirements required for PI status. 28Clarification and awareness of who can and who cannot become a PI could therefore be considered to facilitate clinical trial investigator diversity.
Of the CIs and PIs who were medical specialists, half had more than 10 years of experience in their specialty.With increased time in their profession and the possibility of a greater research track record, an important contributing factor may be their likelihood of securing support for their research from their professional network and competitive grant funding sources.In addition, various barriers to early career PI and CI appointments, aside from limited experience itself, have been described in the literature.Interestingly, scientific research institutes have reported a recent drop in specialist trainees who become PIs due to the branching out of personal interests and pursuits beyond the traditional linear path of academia. 29Furthermore, whilst mentorship is integral to clinical research, compromises to an effective mentortrainee relationship may also serve as a barrier to early career PI appointments; commonly reported are the discrepancies in expectations between PIs and trainees. 30iven the challenges associated with clinical trial recruitment, an approach demonstrated to potentially improve patient recruitment includes the allocation of a trainee principal investigator (TPI) position alongside a PI. 31 Such arrangements may also go on to facilitate mentorship of early career PI appointments and assist with the development of their research track records.
In the period of this study, four out of five and two out of five males have served as TROG clinical trial CIs and PIs, respectively.Additionally, whilst a total of 17 trials adopted the co-CI approach, these only included malemale (58.8%) or male-female (41.1%) arrangements.The disparity in CI and PI appointments between males and females is consistently recorded in other fields of scientific research. 32he lower portion of TROG investigators who are female is likely attributed to the lower number of female ROs (40.7%) and MOs (39.1%) in the Australian workforce compared to males as reported by the Australian Government Department of Health. 33,34However, this alone does not explain the extent of the disparities between female and male investigators.A survey by Martinez et al. 32 involving 1,300 female and male postdoctoral fellows systematically explores the reason for the lower representation of female PIs in scientific research.The survey revealed that over two-thirds of males expressed their interest in wanting to become a PI compared to only one-half of females and that a greater proportion of males would persevere to become a PI compared to females (58% vs. 41%).Moreover, females were less confident that they would be able to attain a PI position compared to males (43% vs. 55%) despite possessing similar baseline professional skills.When it comes to the barriers to becoming a PI from respondents' perspectives, children (21% vs. 7%) and spending time with family (40% vs. 25%) were ranked as a much higher priority amongst females versus males.Notably, perhaps due to their more direct role in caring for dependents, married females (46%) were also most likely to be concerned about the burden of a demanding schedule compared to single females (35%) and males (22%).
Whilst the matter itself is nuanced; the above factors may also help to explain the trends identified in this current study.
The gap between male and female CIs and PIs narrowed during the first two decades of the 21st century.As reported by the Australian Bureau of Statistics, this may be explained by the increasing number of practising Australian female medical specialists during this timeframe. 35Beyond 2020 to the time of this study, however, the number of female CIs and PIs has declined.This may be due, in part, to the greater impact of COVID-19 on females such as the need for home schooling and other caring roles.The systematic review by Pinho-Gomes et al. 36 expounds on this and goes on to reveal that female-only accounted for a third of all authors who published COVID-19-related papers during its outbreak in 2020.Similarly, the study by Batumalai et al. 37 found that female first authorship in the Journal of Medical Imaging and Radiation Oncology (JMIRO) dropped from 38% in 2020 to 28% in 2021.
For females intent on becoming site PI or CI, prominent enablers outlined in the literature include receiving adequate mentorship from senior investigators and colleagues at the site of their work, being supported by family-friendly policies, and feeling empowered by an accommodating work culture. 32hief investigators and PIs have led clinical trials more often in the public compared to the private setting.This may be attributed to the fact that whilst similar numbers of public versus private radiation therapy centres exist in Australia, private departments are on average smaller with a lower proportion of employed ROs (25.8%) compared to the public sector. 33,38Last, whilst generalisations cannot be made for every private department, further barriers to clinical trial initiation in private centres may include busy workflows focussed on treatment service provision and patient throughput, inadequate research infrastructure and lack of internal support for clinical research. 39Partnerships with the private sector and facilitating them in addressing such barriers may be worthwhile to enhance clinical trial activities across private sectors.This will potentially improve DEI in researchers as well as enhance the accessibility of clinical trials and improve DEI in participants.
100% of CIs had initiated TROG clinical trials whilst affiliated with metropolitan institutions and 94.9% of PIs ran TROG trials in metropolitan institutions.Becoming a CI of a clinical trial often requires academic experience and track record, institutional research support and professional and academic networks.Thus, it is not surprising that CIs have all been from metropolitan institutions which are likely to have been tertiary academic centres.The PI predominance in metropolitan centres is also likely to be directly attributed to the higher number of ROs (83.3%) and MOs (83.9%) working within metropolitan areas as well as the disproportionately higher number of radiation oncology departments in metropolitan compared to regional and rural Australia. 33,34,40The Australian Department of Health aims to bridge this gap by providing grants for both public and private sectors to develop radiation therapy treatment centres in regional Australia. 41Given this and given that over one-fourth of individuals reside in regional and rural Australia, much work will be required ahead to ensure equitable clinical trial access and optimal health outcomes for nonmetropolitan populations. 42he retrospective nature of this study carries inherent limitations particularly relating to data quality and missing data (e.g.there were instances when the investigator's year of fellowship was not listed on the AHPRA and MCNZ websites).Methods for identifying missing data included attaining information from the investigator's institution, Linkedin account or personal networks; however, there may have been inaccuracies in these approaches.Furthermore, the start date for data collection between CIs and PIs was different due to PI data unavailability before 2000 when in hospital paper-based systems were used.Last, whilst the Modified Monash Model (MMM) 2019 is the most appropriate system for categorising current geographical locations in Australia, it may not accurately represent the geographical status of institutions at the time of trial opening due to potential urban sprawl over the past five decades.
In conclusion, most of TROG's 134 CIs and 465 PIs have been led by ROs and those with over 10 years of experience.Only one in five trials and one in three trials have been led by female PIs and CIs, respectively.Investigators have largely been affiliated with public institutions, with only one in 100 CIs and one in eight PIs being affiliated with the private sector.TROG members from regional and rural areas in Australia have not been engaged as investigators, with all CIs and most PIs affiliated with metropolitan institutions.Further unpacking and understanding of issues related to CI and PI diversity are important to inform initiatives to improve researcher, leadership and patient diversity in future TROG clinical trials.

Table 2 .
Principal investigator (PI) characteristicsIndividuals (n=) N (% of total number) © 2023 The Authors.Journal of Medical Imaging and Radiation Oncology published by John Wiley & Sons Australia, Ltd on behalf of Royal Australian and New Zealand College of Radiologists.