(J-380) Test-retest Reliability of Neurovascular Coupling Measurements for the Evaluation of Treatment-Related Effects in Cancer Patients

Multimodal functional near-infrared spectroscopy (fNIRS) and electroencephalography (EEG) are approaches to brain imaging used to detect neurovascular coupling (NVC) in humans. EEG records neuronal electrical activity, which requires oxygen for electricity generation. Meanwhile independently, fNIRS can detect hemodynamic responses in the brain from various stimuli by detecting the regional increase and decrease of oxygenated and deoxygenated hemoglobin in the blood. Therefore, there will be a ratio, namely neurovascular coupling, between hemodynamic and electrical responses when commonly responding to a task. The NVC ratio is hypothesized to be more reliable than the response in a single EEG/fNIRS recording alone, at the presence of intra- and inter-session variance. The goal of the current study is to test the reliability of repeated NVC measurements in healthy subjects. The long-term objective is to yield a comparison on NVC measures in subjects with stage I-IV epithelial ovarian or endometrial cancer before and after chemotherapy treatment. The new knowledge on the mechanism of chemotherapy related cognitive impairment in cancer patients can potentially lead to further advancements of brain protection during treatment.

In a group of healthy subjects, fNIRS and EEG recordings were collected during two experimental visits 2-7 days apart. Inclusion criteria are men and women between the ages of 18-59 and have no history of neurological disorders or brain damage. Subjects must also have normal hearing and vision, good use of both hands, and be fluent in English. Each subject completed 6 recording sessions including 2 resting states, and 2 motor tasks. Resting states included 6 minutes of resting with the subject only directed to look at the cross on a computer screen. Motor tasks included 7 blocks of fist clenching for 20 seconds at 1Hz with 30 seconds of resting between. In addition to the healthy subjects, fNIRS and EEG recordings have also been collected in subjects with stage I-IV epithelial ovarian or endometrial cancer. Inclusion criteria are women between the ages of 19-99 and have no history of neurological disorders or brain damage with stage I-IV epithelial ovarian or endometrial cancer. Patients will receive chemotherapy using Carboplatin and Paclitaxel (or Docetaxel in the case of allergies). Patients who are anemic (hemoglobin values less than 11 g/dL) and/or who are being treated for anemia will be excluded. Subjects must also have normal hearing and vision, good use of both hands, and be fluent in English. The same resting state and motor tasks were performed along with a test of attentional control.

Currently, simultaneous fNIRS-EEG data have been collected in 9 healthy subjects and 5 cancer subjects. fNIRS imaging in both healthy and cancer subjects show consistent increase in oxygenated hemoglobin signals and decrease in deoxygenated hemoglobin signals as response to the motor tasks. Furthermore, the fNIRS responses show consistent amplitudes in the repeated measures in healthy subjects, indicating a reliable fNIRS response. Meanwhile, a decrease of oxygenated hemoglobin in fNIRS response has been observed in cancer patients after completing a course of chemotherapy. Further analysis will include calculating the NVC ratio of hemodynamic response to electrical activity in healthy and cancer patients and performing a comparison. In conclusion, our study has demonstrated the feasibility of measuring NVC in healthy and cancer patients. Through continuing the data collection, the project is expected to yield new knowledge on multimodal imaging of NVC, and how NVC may be an indicator of treatment-related effect in cancer patients.