Progression-free survival (PFS) is no longer a useful surrogate endpoint for overall survival (OS) in trials of newer ovarian cancer therapies, Australian researchers have shown.
While correlation between PFS and OS for treatment effect was relatively good in the pre-platinum/paclitaxel era (R2 = 0.66), PFS had a poor correlation with OS in more recent trials of biologicals (0.3) in ovarian cancer, a study at the NHMRC Clinical Trials Centre in Sydney found.
The weakening correlation between PFS and OS over the years was likely due to the impact of more effective salvage therapies diluting the relationship, according to researchers from the Centre’s Australia and NZ Gynaecological Oncology Group.
Writing in Therapeutic Advances in Medical Oncology, they said PFS had been adopted as the primary endpoint in many randomised controlled trials (RCTs) of treatments for epithelial ovarian cancer in the belief that it was a reasonable surrogate for the ‘gold standard’ endpoint of OS.
But in a retrospective review of 26 RCTs involving almost 25,000 patients with ovarian cancer they found the overall correlation between PFS and OS was modest (R2 = 0.52). The strength of correlation diminished over time from 0.66 for trials in the pre-platinum era to 0.44 in the platinum/paclitaxel era, 0.22 for triplet combinations and 0.30 for biological clinical trials.
The researchers noted that correlation in outcomes was weaker in modern trials that had a longer duration of post-progression survival, and this was likely due to the use of more effective second-line salvage therapies.
They expected the relationship between PFS and OS would continue to decline with the increasing availability of treatment options, including crossover to the active experimental treatment following disease progression.
“Therefore, it is increasingly unlikely future trials will demonstrate a relative improvement in treatment effect for OS with first-line therapy. Using OS as primary endpoint will require larger, longer trials in order for first-line treatments to demonstrate an OS benefit.”
The study authors noted that despite advances in treatment ovarian cancer still had a poor prognosis for many women, with most patients with stage III disease relapsing within two years and more than half dying within five years.
The new findings on outcomes would had implications for how trials are designed, they suggested, with a need to have standardised post-progression treatments and adjustment for salvage therapies.
And since PFS was more prone to bias, trials will need to adopt stricter and more consistent definitions of progression.
With the value of PFS as a primary endpoint being the subject of much debate, it may need to be backed up with other endpoints such as patient reported outcomes and time to second disease progression, the researchers said.
“Alternatively endpoints such as quality-adjusted PFS, which represent a measure of net clinical benefit, could be used as primary endpoints and for clinical decision making and regulatory approval.”
“It is also important to demonstrate no OS detriment if PFS is used as the primary endpoint,” they concluded.