Rapid Fire Best of the Best Poster Oral Clinical Oncology Society of Australia Annual Scientific Meeting 2022

A randomised controlled trial evaluating two cognitive rehabilitation approaches for cancer survivors with perceived cognitive impairment   (#119)

Janette L Vardy 1 2 , Gregory R Pond 3 , Melanie L Bell 4 , Corrinne Renton 5 , Ann Dixon 5 , Haryana M Dhillon 5
  1. University of Sydney, Concord, NSW, Australia
  2. Concord Cancer Centre, Concord, NSW, Australia
  3. McMaster University, Hamilton, Ontario, Canada
  4. University of Arizona, Tuscon, USA
  5. CeMPED, University of Sydney, Sydney, NSW, Australia

Background:

Up to 70% of survivors report cognitive symptoms after chemotherapy.  We compared two cognitive rehabilitation programs to a control group in cancer survivors.

 

Methods:

Study population consisted of adult cancer survivors with persistent cognitive symptoms 6-60 months after adjuvant chemotherapy.  Participants randomised to: Attention Process Training [APT], Compensatory Strategy Training [CST], or control group. Active interventions comprised 6-week, 2-hour/week small-group sessions. Assessments: pre- and post-intervention, 6- and 12-months later. Focus groups/interviews held after initial treatment groups. Primary outcome was change in self-reported cognitive function (Functional Assessment of Cancer Therapy-Cognitive Function, Perceived Cognitive Impairment [FACT-COG-PCI] subscale) between baseline and post-intervention. Secondary endpoints included: objective neuropsychological performance, Functional Impact Assessment (FIA), Patient-Reported Outcome Measures, and associations between variables.  Analyses were on an intention-to-treat basis. Analysis of covariance mixed models were used to model continuous outcomes.

 

Results:

Sixty-five participants were randomised (APT n=21; CST n= 24; Controls n= 20): 94% breast cancer, 98% female, median age 54. Median time since chemotherapy 20.7 (range 8-60) months. FACT-COG-PCI improved in all groups over time, but no significant differences were observed between arms. Likewise, for mean clinical neuropsychological T-scores and FIA no difference between intervention groups or time*intervention was seen. On mean clinical neuropsychological T-scores 19/65  (29%) were impaired at baseline; post-intervention impairment Controls 31.3,%, CST 16.7%, and APT 20.0%. On FIA at baseline nine were impaired (Controls 10%; CST 17.4%; APT 15.8%); this decreased to three post-intervention (one/group).

FACT-COG-PCI was weakly associated with clinical neuropsychological tests (rho=0.24, p=0.051) at baseline, and had no association with FIA. Neuropsychological total mean T-score was moderately positively associated with FIA (rho=0.37, p=0.003).

Conclusion:

There were no significant differences between intervention groups and controls using linear mixed models adjusted for baseline scores. An improvement was seen in cognitive symptoms and neuropsychological test scores over time in all groups.