Biomechanical effect of neurologic dance training (NDT) for breast cancer survivors with chemotherapy-induced neuropathy: study protocol for a randomized controlled trial and preliminary baseline data

We measure postural dynamics as postural control or sway. The primary outcome, postural control variability, is measured through biomechanical data collection during a challenging 30-s balance task, that has been described previously [47, 56]. Briefly, participants are instructed to stand quietly and bilaterally for 30 s with eyes closed (QEC) on a portable balance plate (Bertec Corporation, Columbus, OH), with feet set apart (5 cm), and arms relaxed at their sides. To control for the effect of auditory stimulation [32], the test is performed in silence within a quiet environment. To control for known responses to vision obstruction, participants are asked to focus on a point approximately 8 ft in front of them at eye level prior to closing their eyes at the moment collection commenced [4, 9, 43, 47, 56] and in response to the researcher's countdown of “3.0.2.0.1..close.” Biomechanical data describing postural sway are acquired using custom software written in LabVIEW at a sampling frequency of 1000 Hz and consist of one channel of force data (Fz) and two channels of moment data (Mx, My) over time. From these data, we calculate the COP time series for the 30-s duration of QEC condition performance. We then calculate the primary outcome measures of COP variability in resultant and medial–lateral planes (RMSr, RMSml) for this 30-s duration per Prieto [57].

We measure gait dynamics as gait variability [33, 51, 68] and stability [69‐71] using data from inertial measurement units affixed to the foot, leg, thigh, pelvis, and/or lower cervical spine during the 6-min walk test (6mwt). The primary outcome measure for Aim2, CVspeed, will be calculated from acceleration data [72]. Steady-state walking epochs are detected from gyroscopic data.

Starting from the QEC COP time series data described in postural dynamics (primary outcome measure), we calculate additional linear [57] and non-linear [58] secondary outcome measures. In addition to linear COP primary outcome measures (1) sway variability resultant (RMSr), (2) sway variability in the frontal plane (RMSml), and (3) variability in the anterior–posterior plane (RMSap), we calculate secondary outcome measures (4) 95% ellipse area (COPa), (5) path length (PL), and (6–8) mean velocity in resultant, medial–lateral, and anterior–posterior planes (COPvr, COPvml, COPvap). We analyze one non-linear COP measure: sample entropy of the resultant COP position using the increment calculation method (SEI) with constant values applied of m = 3 and r = 0.3 [58]. Entropy of postural responses represents the automatic complexity [73] of neuromotor control that is available to an individual and has been found sensitive to disease states such as brain injury [74] and neurodegenerative disease [75‐78] as well as to skill mastery [79] and attentional focus [58, 80, 81]. Within this manuscript, we report descriptive statistics for these primary and secondary outcomes describing postural control and for select clinical measures of function among BC survivors enrolled to date (n = 23). Within-subject variability at baseline as well as intervention effects will be reported in future publications.

We measure function using the following validated clinical tests, instrumenting some of them to accomplish biomechanical quantification of the motion performed during the tests. All functional tests are performed in a clinical space or the participant’s home in an area that is distraction free. The Timed Up-and-Go test (TUG) is a timed test of a person’s ability to stand from a chair, walk 10 feet, turn around, and return to sitting [82] with shorter times indicating better functional balance. Presence of CIN has been associated with longer TUG times [11‐14]. To measure dual-task function, we use the TUG performed simultaneously with a cognitive task consisting of audibly counting backward by 3 s or 4 s (TUGCog). Blackwood and Rybicki (2021) reported that the TUGCog threshold of ≥ 11.32 s identified BC survivors at risk of falling (sensitivity = 0.64 and specificity = 0.8) [13]. To measure dynamic balance function we collect Mini Balance Evaluation System Test (MiniBEST), which evaluates sensory organization, anticipatory and reactive postural control, and dynamic gait indices [5] and was found to discriminate BC survivors from controls in at least 1 prior study [12]. Finally, to measure physical function in terms of endurance and mobility we collect 6mwt, in which shorter distances correlate with poorer function including among survivors with CIN [83]. Meta-analysis of 6mwt results among adults with pathology or fear of falling indicate that 14.0–30.5 m represents the minimum clinically-important difference (MCID) for improvement in this measure [84]. The 6mwt is performed within an oncology clinic, in a quiet area without distractions. Participants walk in a loop with the following configuration: a straight distance of 65 feet (ft) minimum marked by tape on the floor that participants turn around to their left; participants are cued to walk straight between tape marks and to turn comfortably at the end of each straightaway at the fastest pace they feel they can maintain for 6 min, resting as needed in a standing position before continuing.

Measures of walking variability and stability have been found indicative of mild cognitive impairment [85], age [86], fall risk [51, 68, 87], and neuropathy [88, 89] regardless of age or type of cancer [90] including specifically in BC [10, 89, 91]. Using a portable inertial measurement unit (IMU) system while participants perform the 6mwt [92], stride-to-stride measures of variability previously found sensitive to health status will be calculated from the straight, steady-state sections of the 6mwt loop pattern [46, 51, 68]. Local dynamic stability [93‐96] and orbital stability [10, 96, 97] will also be analyzed, as measures that have been shown sensitive to neuropathy and used previously by our team to measure intervention effect among BC survivors [98].

is a ratio of the activity produced in agonist v. antagonist muscles of the leg that provides unique insight into neuropathy effects [99]. We calculate this ratio from electromyographic (EMG) signals of the tibialis anterior, gastrocnemius, and soleus muscles. We calculate this muscle activity ratio during performance of select MiniBEST sensory organization tests as well as during QEC and QECm postural dynamics trials and the 6mwt.

We measure the influence of music listening on postural dynamics (QECm) per the protocol established by Ross et al. [32]. After completing the QEC trial (and always in this order), the participant is asked to step off the balance plate and walk or march in place for a few steps before being asked to step back on the plate into the QEC posture while we begin playing the musical composition "La Cumparsita" at a volume that is deemed loud but not uncomfortable by the participant. Participants are instructed to look at a defined spot approximately 8 feet away, during approximately, and no less than, 10 s of acclimating to auditory stimulus, before being cued to close their eyes for collection of the QECm trial. Auditory acclimation is accomplished in the following way: we instruct participants to listen to the music, focusing specifically on the periodic rhythm (i.e., “beat”). To facilitate beat perception, we provide external cueing (i.e., snapping fingers or clapping hands to the musical beat) for approximately 4 s. We then extinguish external cueing over a 2-s period, by quieting then ceasing the cueing sound. Then participants are allowed to acclimate to auditory stimuli, without external cueing, for an additional 4 s before research staff verbalize the directions “3.0.2.0.1..close,” speaking one word per beat for a measure of 4 counts and start of postural control data recording as participants close their eyes. COP-based variables of interest are calculated as described for the QEC trial.

Patient self-report surveys are administered through The Ohio State University administered instance of the REDCap database for in-person visits and through the REDCap platform’s MyCap mobile application (MyCap) for data entry that participants perform remotely (e.g., HEX activity, fall tracking in 6-month follow-up). Answers are reviewed manually by authorized research collaborators, who prompt participants to complete or clarify answers as needed and document their responsibility for review from a list of authorized data reviewers. Within this study, we collect the following PRO measures: (i.) European Organization for Research and Treatment of Cancer’s Quality of Life Questionnaire, Chemotherapy-Induced Peripheral Neuropathy (CIPN-20): is a validated 20-item patient-reported questionnaire instrument for longitudinal evaluation of neuropathy symptoms induced by chemotherapy(108). While validated for a recall period of “during the past week” we use modified forms of the CIPN-20 to query items for the recall periods of “during the activity just performed” (at the end of sessions) and “right now” (at the beginning and end of sessions). (ii.) The Brief Pain Inventory (BPI) is validated for ecological momentary assessment (EMA) of immediate and retrospective (prior 24 h) self-reported pain and relevant functional capacity in our target population [100]. (iii.) The Brief Fatigue Inventory (BFI) is validated for EMA of immediate and retrospective (prior 24 h) self-reported fatigue and relevant functional capacity in our target population. (iv.) Ecological momentary assessment (EMA) of immediate symptomatology. Using the Patient-Reported Outcome version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE), cancer-related symptoms such as balance problems, nausea or vomiting, dizziness, sensitivity to light or noise, feeling like “in a fog”, confusion, sadness, and anxiety can be queried for the recall period of the past 7 days [101]. This study requires query of symptomatology in the immediate moment, or “right now”: we capture these symptoms in this timeframe using an EMA tool that prompts for self-report of symptom severity “right now” called the Sports Concussion Assessment Tool (SCAT) symptom inventory [102]. The SCAT was created for and validated among individuals with mild traumatic brain injury, however, the SCAT symptom inventory queries the desired PRO-CTCAE items (i.e., 39–56) in the time frame of “right now” using a 7-point Likert scale (0 = no incidence; 6 = high severity), which affords greater sensitivity than the PRO-CTCAE 4 point Likert scale (1 = no incidence; 4 = high severity), and has been validated for use among individuals with known cognitive deficits. Therefore, we use the SCAT symptoms inventory in lieu of the PRO-CTCAE to query the desired subset of the PRO-CTCAE symptoms with regard to how participants feel “right now” with sensitivity and in a format appropriate for individuals who might have cognitive deficits associated with chemotherapy exposure. (v.) Activity tracking: Participants are asked if they have participated in the following activities: physical therapy, occupational therapy, fitness activity, other therapeutic or fitness activity. If any category of activity is checked as having occurred then participants are prompted to estimate the amount of time spent doing the activity since we last saw them (hours, minutes). (vi.) Falls tracking: Falling is defined as an unexpected loss of balance in which an individual comes to rest at a position lower than before the unexpected event [103]. We elicit self-report of falls and loss of balance using the question “How many times have you fallen or felt like you lost your balance since we last saw you?” Responses are typed by the participant and reviewed for incidence of falls, incidence of loss of balance, and details offered about either incidence. (vii.) Barriers to participation: Participants are asked “To help us understand barriers to participation, if you missed a session, please indicate the reason why you were unable to attend (e.g., schedule conflict, transportation, didn’t feel up to it, forgot).” (viii.) Satisfaction with intervention is measured after each class using a 7pt Likert scale and prompt for feedback about what did/did not work per class. Feedback is used to improve future sessions. (ix.) The Intrinsic Motivation Inventory (IMI) was developed from the perspective of Self Determination Theory to assess 7 dimensions of experience: Interest/Enjoyment, Perceived Competence, Effort/Importance, Pressure/Tension, and Choice [104]. We administer the 9-item short form of the IMI monthly to optimize instruction around low Pressure/Tension and high Interest/Enjoyment [61, 62] as well as to explore relationships between adherence, motor effects, and IMI dimensions including perceived benefit (i.e., Effort/Importance).