Bladder cancer patients who were ctDNA negative before receiving adjuvant atezolizumab experience more immune-related adverse events (irAEs) without an apparent survival benefit, researchers report.
The post hoc analysis, funded by Roche/Genentech and co-authored by company representatives, found that patients with muscle-invasive urothelial carcinoma (MIUC) who were ctDNA negative at baseline had nearly double the risk of low-grade irAEs with adjuvant atezolizumab compared to ctDNA positive patients (HR 1.91), while rates of severe events were similar between the groups.
By contrast, only patients who were ctDNA positive seemed to derive an overall survival benefit from adjuvant atezolizumab versus observation (HR 0.58), whereas no survival benefit was observed in those who were ctDNA negative.
The study “provides a rationale for further exploration of ctDNA as a predictive marker for identifying who to treat and when in an individual’s cancer journey in order to ensure a favourable benefit/risk balance,” wrote the authors, led by Professor Thomas Powles, Director of the Barts Cancer Centre at St. Bartholomew’s Hospital.
The study, published in ESMO Open [link here], analysed MIUC patients who had undergone radical cystectomy and were treated in the IMvigor010 trial with either adjuvant atezolizumab (n=384) or observation (n=397).
The proportion of patients who were ctDNA-positive at baseline was similar in both the atezolizumab and observation arms (39% vs 35%).
Among patients treated with atezolizumab, any-grade irAEs were more frequent in those with negative baseline ctDNA than those positive for ctDNA (55% vs 33%).
Grade 1-2 irAEs were particularly frequent in ctDNA-negative patients (50% vs 29% in ctDNA-positive), with rash (28%), hypo-/hyperthyroidism (19%) and hepatitis (16%) most common.
“The higher rate of irAEs due to thyroid dysfunction in patients who received atezolizumab and were ctDNA-negative at baseline was more than twice that of patients who were ctDNA-positive (19.3% versus 7.9%, respectively),” the authors noted.
However, rates of grade 3-5 irAEs were similar in both ctDNA-negative and -positive patients (10% vs 8%).
Baseline ctDNA-negativity was associated with a higher risk of low-grade irAEs, and the odds of low-grade irAEs fell as baseline levels rose, with an 8% decrease in risk per increase in ctDNA levels by 1 log-MTM/ml unit.
The risk of high-grade irAEs was similar for both ctDNA-negative and -positive patients, however, the authors noted that there were very few events.
There was no association between irAEs and on-treatment changes in ctDNA status, as patients who were ctDNA-positive at baseline but ctDNA-negative by day 43 showed no increased or decreased risk of irAEs compared to patients who remained ctDNA-positive.
The authors noted that the findings were limited by the nature of a post-hoc analysis and that they should be confirmed in prospective studies.
“Further work is needed on understanding the sensitivity and specificity of monitoring ctDNA at the level of the individual test and the ability of each test to accurately detect ctDNA, as well as the suitability of ctDNA monitoring across tumour types, given the different rates of DNA shedding by different tumours,” they added.