Patient-reported outcomes (PRO) in ovarian cancer clinical trials—lost opportunities and lessons learned

open access

 Despite increased recognition of the value of including patient-reported outcomes (PROs) as important end points in phase III clinical trials, there has been a lack of pre-specified PRO hypotheses and shortcomings with the analyses and interpretation of PROs in many ovarian cancer trials. This paper discusses and provides examples of the so-called lost opportunities in ovarian cancer trials.

 In ovarian cancer clinical trials, the diagnosis of progression is based on standardised RECISTv1.1/GCIG CA125 criteria, which then determine whether the patient remains on treatment or comes off the trial. It is unclear how many patients diagnosed with ‘progressive disease’ have any symptoms that could be attributed to progression, as this information is seldom, if ever, reported.

 It is beyond the scope of this paper to discuss all the problems and vagaries of RECIST v1.1. The assumption underlying RECIST is that tumour measurements can be reliably carried out by different readers and are accurate and reproducible. However, this is open to question [2, 3]. The reliability and reproducibility of RECIST v1.1 in ovarian cancer is unknown. Furthermore, recurrent ovarian cancers can be very difficult to ‘measure’ on a computed tomography scan. It has been estimated that over 50% of patients with recurrent disease do not have measureable disease using RECIST criteria [4]. We are entering a new era in which many patients with ovarian cancer will be treated with targeted therapies and immune checkpoint inhibitors which are quite different from cytotoxic chemotherapy and this raises questions regarding the value of using RECIST to measure benefit with other treatment modalities. These problems with RECIST v1.1 and the somewhat arbitrary definitions of progression underscore the importance of including patient-reported outcomes (PROs) as end points in clinical trials.

 conclusion

There is increasing recognition regarding the potential importance of including PROs (patient reported outcomes) in ovarian cancer clinical trials. The Society of Gynecologic Oncology (SGO) white paper on clinical trial end points concluded that end point selection in ovarian cancer clinical trials should reflect the impact on disease burden and unique characteristics of the treatment cohort while reflecting true patient benefit [26, 27]. The 4th Gynaecologic Cancer Intergroup (GCIG) consensus meeting also recognised the importance of HRQOL in all clinical trials and concluded that clinical trials in patients with platinum-resistant/refractory ovarian cancer should at the very least measure the impact of treatment on HRQOL and symptom benefit. Ideally, these should be used as co-primary end points with traditional end points such as PFS and OS [26, 28]. For this to be made possible, regulatory bodies must follow the lead of ESMO and MCBS and recognise that there are other meaningful ways to measure treatment benefit apart from the impact on PFS or OS. We could have learned much more from previous ovarian cancer trials if there had been closer attention to PRO hypotheses and the inclusion of appropriate PRO end points when designing the clinical trials. Ongoing initiatives promoting the value and importance of high-quality PRO data in clinical trials will hopefully turn the tide and bring about change.