While incident cancer cases of any site remain stable, 5-year cancer survival rates have increased [1]. In January 2022, >18M individuals in the US were cancer survivors—defined here as those diagnosed with cancer onward [2]—with a projected increase to 26M by 2040 [3]. Although modern cancer treatments are effective, these treatments can worsen cancer survivors’ health-related quality of life (HRQoL) [4-6]. Several cancer treatments—most notably, chemotherapy—are associated with cognitive difficulties [7-9]. While chemotherapy effectively treats many malignancies, it also negatively impacts vascular endothelial cells [10-12]—leading to pathophysiological processes contributing to cancer survivors’ often-reported chemotherapy-induced cognitive impairment (CICI; aka “chemo-brain”) during/after treatment [7,8]. As the number of cancer survivors increases, CICI mitigation research is crucial to improve HRQoL.

CICI likely occurs for several reasons. Chemotherapy can induce immune and antioxidant dysregulation [13-16]. This dysregulation can lead to peripheral proinflammatory cytokines being released, which damage and cross the blood-brain barrier [13,14,17]—contributing to central proinflammatory cytokine release [18-21] and causing neuroinflammation-related impairments in the myelination process [22]. Endothelial dysfunction and cerebral autoregulation failure can also result from this damage [12,23-25]. Reviews [10,11] suggest these mechanisms alter cerebrovascular function more broadly by impairing cerebral perfusion, glucose metabolism, and angiogenesis; thus contributing to CICI.

Type 2 diabetes mellitus (T2DM) is both a risk factor for developing several common cancers and, given its prevalence, a common cancer survivor comorbidity [26,27], heightening cancer recurrence risk [28,29]. Several pathophysiological mechanisms underlying CICI are observed with T2DM [30]. Insulin resistance can result in repeated hyperglycemic episodes and subsequent proinflammatory cytokine elevation [31-35]—possibly explaining why insulin resistance is correlated with diminished cerebrovascular and cognitive function in individuals with T2DM but without cancer [36-38]. This might also explain why, in newly diagnosed cancer survivors with proinflammatory comorbidities (eg, T2DM), higher proinflammatory cytokines and poorer cognitive function have been observed relative to cancer survivors without comorbidities—even prior to chemotherapy [39]. CICI mitigation research that compares those with and those without T2DM may help clarify the magnitude to which T2DM contributes to the mechanisms underlying CICI. Notably, CICI mitigation research may also benefit from the inclusion of assessments of the epigenome, given that upregulated proinflammatory cytokines, a hallmark of chemotherapy and T2DM, may lead to changes in DNA methylation via histone modification [21,40,41].

Aerobic physical activity (PA) and resistance training (RT) have robust health benefits counteracting much of the pathophysiology underlying CICI and T2DM [42,43]. Reviews [44,45] suggest better cerebrovascular function in those who are more active or fitter relative to those who are less active or fit, with greater fitness positively associated with cognitive function in other at-risk populations [46,47]. In those with T2DM, research has shown that PA is impactful at lowering proinflammatory cytokines [42,48-50] and improving insulin sensitivity [51,52]. Yet, research is needed on how increasing PA in cancer survivors with T2DM reporting CICI improves cerebrovascular and cognitive health through improved fitness. Notably, cancer survivors prefer supportive, individualized, home-based PA programming that includes RT and aerobic activities such as walking [53]. Home-based programs are recommended by the American Cancer Society [54] given their ability to circumvent some of the barriers to PA program participation (eg, distance to clinic, lack of time) among those living with and beyond cancer. Smartphone app technology can assist in providing this home-based PA program access. Relative to routine care, smartphone app–based supportive care among cancer survivors has been observed to improve HRQoL, self-efficacy, and symptoms of distress and depression [55]. Further, multicomponent interventions that include personalized PA programming delivered via smartphone apps, with wearable devices to supplement behavior change, have shown promise for increasing PA among cancer survivors [56,57].

We are conducting a 30-participant pilot study in cancer survivors with T2DM (n=15) and without T2DM (n=15)—all self-reporting “chemo-brain”—that includes personalized PA programming and education delivered via 2 smartphone apps, with a wearable device provided for further behavior change support. We have detailed our Specific Aims and Hypotheses below.

Characterize cerebrovascular function, cognitive function, proinflammatory cytokines, and cardiometabolic outcome differences between cancer survivors with and without T2DM. We hypothesize that cancer survivors with T2DM will have (1) poorer measures of middle cerebral artery velocity and cerebrovascular conductance, reactivity, and resistance; (2) lower executive function scores; (3) higher C-reactive protein, interleukin-6, and monocyte chemoattractant protein 1 concentrations; and (4) higher blood pressure and insulin resistance.

Examine pre- to post-PA program differences for changes in Specific Aim 1’s outcomes between cancer survivors with and without T2DM. We hypothesize that greater pre- to post-PA program improvements in these outcomes will be observed for cancer survivors with T2DM.

Assess pre- to post-PA program differences for changes in Social Cognitive Theory–based (SCT-based) [58,59] constructs, select psychosocial outcomes, HRQoL, and PA between cancer survivors with and without T2DM. We hypothesize that cancer survivors with and without T2DM will demonstrate pre- to post-PA program improvements in (1) SCT-based, PA-related self-efficacy, outcome expectations, enjoyment, social support, and barriers; (2) stress, anxiety, mood, and depressive symptoms, and that cancer survivors with T2DM will have greater improvements than cancer survivors without T2DM; (3) in HRQoL due to greater improvements in physical functioning as well as pain intensity/interference, and (4) overall PA participation due to greater increases in light and moderate PA.

Characterize epigenetic profiles of biologically relevant pathways related to neurocognition, energy metabolism, and proinflammatory processes in cancer survivors with and without T2DM, and whether changes are observed in these profiles pre- to post-PA program. We hypothesize that distinct epigenetic profiles will be evident between groups but that both groups will show pre- to post-PA program changes in these profiles.

We are conducting a National Institutes of Health Stage 1 trial [60], and we have registered this trial on ClinicalTrials.gov (NCT06725953). Moreover, we used the SPIRIT (Standard Protocol Items: Recommendations for Interventional Trials)-Outcomes Guidelines to guide the construction of this manuscript [61]. We are recruiting individuals within Oklahoma City and the greater state of Oklahoma. Our accrual target is 38 participants to allow our desired number of 30 participants to be achieved when accounting for an anticipated 20% attrition rate [62]. We are conducting rolling recruitment and are recruiting via (1) emailed and posted flyers; (2) digital advertising to local survivorship groups; (3) oncologist referral; and (4) chart reviews/assistance from the Stephenson Cancer Center Clinical Trials Office. Potentially eligible individuals are sent a REDCap-based (Research Electronic Data Capture–based) screening survey, and, whenever possible, we are verifying cancer diagnosis, chemotherapy regimen, and T2DM status via medical records. We are evaluating individuals against stringent eligibility criteria. Specifically, our inclusion criteria are as follows: (1) 18 years or older; (2) ability to speak/read English; (3) ability to provide informed consent; (4) underwent treatment within the last 3 years for a non–brain-related cancer with said treatment having included chemotherapy and not currently undergoing any active cancer treatments (ongoing endocrine/hormonal therapy or immunotherapy, when either are used for maintenance purposes, are acceptable); (5) self-reported cognitive difficulties following cancer treatment; (6) (for cancer survivors with T2DM) current T2DM diagnosis as classified by a fasting blood glucose of ≥126 mg/dL, 2-hour oral glucose tolerance test of ≥200 mg/dL, HbA_1c level of ≥6.5%, or use of medications (eg, Metformin) or insulin to treat hyperglycemia or (for cancer survivors) no current T2DM diagnosis as classified by a fasting blood glucose <100 mg/dL, 2-hour oral glucose tolerance test <140 mg/dL, or HbA_1c level of <5.7%. Presence/absence of T2DM to be confirmed preferentially via physician’s documentation or via our measurement of participants’ fasting blood glucose during the blood draws within Part 2 of Baseline Testing (Figure 1); (7) own smartphone or computer with internet access; and (8) willing to participate in the 12-week remotely-delivered PA program. Our exclusion criteria are as follows: (1) reporting a Physical Activity Readiness Questionnaire score indicating PA may be unsafe unless a doctor’s note clearing a prospective participant for study participation can be provided; (2) engaging in ≥75 min/week of vigorous-intensity PA, ≥150 min/week of moderate-intensity PA, or an equivalent combination of both over the last 3 months; (3) currently undergoing active cancer treatment or planning more treatment within the next 3 months; (4) classified within the prediabetes range for their fasting blood glucose measurement (100-125 mg/dL) during Part 2 of Baseline Testing. Notably, those with T2DM verified via a physician’s documentation and who report the use of medications or insulin to treat hyperglycemia, who are found to be in the prediabetes range, will still be enrolled within the T2DM group and study. These results would indicate a participant with well-controlled T2DM, not prediabetes; and (5) currently a prisoner, pregnant, or planning to become pregnant during the study.

We will terminate a participant’s participation if (1) the participant desires to voluntarily withdraw from the study or (2) the investigators feel a participant is unable to complete the study protocol due to changes in health status. All adverse events are being tracked within a REDCap tool that meets Common Terminology Criteria for Adverse Events 5.0 requirements. Adverse events are reviewed by the study’s lead investigators (ZCP, MK, and PM) and the study’s oncologist (CH) and reported to the institutional review board (IRB) as necessary (full protocol available upon reasonable request). We do not currently have a Data and Safety Monitoring Board for this study.

We will first inspect all data for outliers/errors and normality, with analyses done in SPSS (version 29; IBM Corp) and R (R Foundation), depending on the metrics being analyzed. Baseline differences between cancer survivors with and without T2DM for (1) sociodemographics, (2) health characteristics, and (3) Aim 1/Exploratory outcomes (ie, cerebrovascular and cognitive function, cardiometabolic outcomes, proinflammatory cytokines, epigenome) will be compared using t tests or chi-square tests, as appropriate. Repeated measures ANOVA will analyze between-group differences for changes in Aim 1/Exploratory outcomes during the PA program (Aim 2) and Aim 3 outcomes (ie, SCT-related constructs, psychosocial outcomes, HRQoL, PA). If any sociodemographic or health characteristics differ at baseline between cancer survivors with and without T2DM, we will use repeated measures ANCOVA. Consistent with recommendations for pilot research [97,98], no a priori power analyses have been performed. Indeed, this study is designed to generate the data necessary to develop larger, properly powered, and federally funded studies on this critical topic.

This study was approved by the University of Oklahoma Health Campus Institutional Review Board (IRB #17332). Participants consent to the study prior to engaging in any study activity. Further, participants can opt out of participation at any point during the study period. All participant data is deidentified, with data accessible to only those on the study team. Participants are compensated with a US $132 gift card upon completion of all study tasks, covering their time participating and travel, while also being able to keep resistance bands and a Fitbit Charge 6.

All data are being securely stored in REDCap and on encrypted University servers. As this is a Multiple Principal Investigator (MPI) study, the MPIs are jointly responsible for monitoring the safety environment of participants and ensuring that appropriate medical care and coverage are provided to all participants, if necessary. The safety monitor (study oncologist; CH) has also pledged to oversee participant safety (eg, reported adverse events). The MPIs are specifically responsible for monitoring all procedures during the conduct of the study for each participant, including eligibility, enrollment, data collection, evaluation of study outcomes, problems with informed consent, and participant safety and well-being.

We are investigating (1) whether cancer survivors with T2DM have differential cerebrovascular, cognitive, proinflammatory, cardiometabolic, and epigenetic profiles versus cancer survivors without T2DM and (2) if a 12-week SCT-based [58,59], technology-delivered PA program results in differential between-group changes in these outcomes as well as SCT-related constructs, select psychosocial metrics (stress, anxiety, mood, etc), HRQoL, and PA. Compared to cancer survivors without T2DM, we expect cancer survivors with T2DM to have poorer measures of cerebrovascular, cognitive, proinflammatory, and cardiometabolic outcomes, as well as distinct epigenetic profiles, during Baseline Testing. However, we expect cancer survivors with T2DM to demonstrate greater improvements in these outcomes during the 12-week PA program relative to cancer survivors without T2DM. Finally, we expect that both groups will demonstrate similar improvements in SCT-related constructs and select psychosocial metrics pre- to post-PA program, but that, compared to cancer survivors without T2DM, cancer survivors with T2DM will have greater improvements in HRQoL and PA.

This study has several strengths. First, we are using an expansive methodology (eg, TCD, venipuncture, saliva collection) and a team science-informed approach to address an important mechanistic question applicable to a high proportion of cancer survivors who have T2DM, or may be at risk for T2DM [26,99], and are currently experiencing cognitive difficulties following cancer treatment. Second, our use of behavior change theory to inform the components and implementation of a PA program personalized to each participant has merit and should help increase the likelihood of the program’s success. Third, and relatedly, the use of technology (2 smartphone apps, a wearable device) to deliver the PA program is notable, as study observations would have implications for how to best provide supportive care to cancer survivors who do not have readily available access to supportive care posttreatment (eg, cancer survivors living in nonurban areas).

Despite the study’s strengths, a few limitations must be considered. First, TCD cannot image the middle cerebral artery diameter; thus, we are holding the middle cerebral artery diameter constant in this study’s analyses. This reduces the ability to measure absolute middle cerebral artery blood flow velocity. However, our test of cerebrovascular reactivity to CO₂ has a strong relationship (r=0.85) with changes in cerebral blood flow [100] and, therefore, helps to partially overcome this limitation. Second, we are recruiting a relatively small sample with limited power to detect significant differences between groups, and, relatedly, there is an absence of a true control group, which may limit the ability to discern true intervention effectiveness. However, the a priori purpose of this study was to gather the data necessary for a future, larger, and fully-powered randomized trial—a trial that would be able to concurrently address both of these limitations. Finally, we are currently recruiting individuals diagnosed with a broad range of cancer types who were treated with chemotherapy. This strategy aligns with the stated study aims and enhances the feasibility of completing this study in a timely manner. Yet, we acknowledge the need to consider specific cancer types that may most benefit from this type of program and, further, how other treatments (eg, radiation) may also be contributing to cognitive difficulties among cancer survivors.

We expect the observations made in our study, particularly during the PA program underpinning Aims 2 and 3, to elucidate unique insights regarding the cognitive benefits of PA at the mechanistic level among cancer survivors, particularly those with T2DM. We want to expand this program to include not only a multisite investigation of how a well-defined PA program in cancer survivors with T2DM improves the outcomes discussed, but also how integration of intervention components regarding other health behaviors (eg, proper sleep, nutrition) might provide additional benefits. For example, integrating a low-touch sleep component into this multisite investigation would be prudent given that up to 75% of cancer survivors experience sleep difficulties [101], which has deleterious impacts on cognitive function and HRQoL while also increasing daytime fatigue—the latter lessening the likelihood of PA participation.