2.1. CRF

The evaluation of the CRF was one of the main criteria for selecting eligible studies for review. CRF was, therefore, part of all nine studies [18,19,20,21,22,23,24,25,26], with four studies having the CRF as the primary outcome [20,21,23,24]. Oxygen consumption (VO₂peak or VO₂max) [18,19,20,21,22,23,26] was used in seven studies to evaluate CRF, of which the VO₂peak/six minute walk test (6MWT) combination [23] was used once and the VO₂peak/treadwalk test was used once [26]. One study further evaluated CRF using the steep ramp test and exercise test at 70% of maximum workload [24] and one study used only the 6MWT [25].

In four of the nine studies, improvement in CRF was reported in the HB exercise group [18,20,22,23]. Husebo et al. [25] further reported a significant difference from baseline and six months post-treatment for total sample (home-based exercise and usual care). Significant improvements were not reported in the other four studies [19,21,24,26].

A total of eight studies compared CRF results between HB exercise and usual care groups [18,20,21,22,23,24,25,26]. Five of these studies reported a significant CRF improvement in the HB intervention [18,21,22,23,26], and one study reported that the effect ceased to be significant after six months of follow-up [18]. The remaining three studies found nonsignificant differences between groups [20,23,25]. Cornette et al. [23] found a significant difference in per-protocol analysis (not intention-to-treat). Two studies also compared HB and CB exercise interventions [19,24]. Whereas Alibhai et al. [19] found nonsignificant difference between interventions, van Waart et al. [24] reported improved CRF results for the CB in comparison with the HB exercise intervention.

2.2. Physical Activity (PA)

A total of seven studies compared the PA levels [18,20,21,23,24,25,26]. Six studies used International Physical Activity Questionnaire [20,21,23,25], Physical Activity Scale for the Elderly [24], and Seven-day Physical Activity Recall [26] questionnaires for evaluation. One study used an accelerometer for PA evaluation [18].

Five of the seven studies reported a significant improvement in PA levels with HB exercise in the intervention group [18,20,23,26], and one study, in the total study sample [25]. In two studies, these results were not reported [21,24]. Cornette et al. [23] reported that the HB exercise group doubled its total PA, but did not describe statistical significance in this context. McNeil et al. [18] reported, in addition to a significant increase in PA levels (both intervention exercise groups), participants reduced sedentary time in the lower-intensity exercise group.

Two of the seven studies further described a statistically significant increase in PA with HB exercise compared with the usual-care group [18,26]; however, Pinto et al. [26] added that group differences attenuated over time. Lahart et al. [21] stated that the effect of the intervention was likely beneficial (91%) on total and moderate PA levels compared with usual care. Another three studies reported a nonsignificant difference between HB exercise and usual care [20,24,25].

2.3. Fatigue and HRQOL

The fatigue outcome was measured in five studies using the Functional Assessment of Cancer Therapy–Fatigue [19,26], Multidimensional Fatigue Inventory [23,24], Fatigue Quality List [24], and Schwartz Cancer Fatigue Scale [25] questionnaires. Reporting fatigue results was insufficient in all the studies. Only one study reported results with statistically nonsignificant difference in the HB group [26]. Three studies reported a nonsignificant difference between the HB exercise and usual-care groups [23,25,26]. An exception is the study by van Waart et al. [24], where a significant reduction in fatigue was found in the HB group, but these results were obtained from a subscale of the questionnaire primarily focused on quality of life. A comparison of average scores at baseline and six months post-treatment further showed a nonsignificant difference, suggesting a return to baseline levels of fatigue [25]. One study [24] reported significantly better physical fatigue results in the CB than HB exercise interventions. Alibhai et al. [19] found nonsignificant important changes in fatigue between interventional groups in HB and CB exercise interventions.

A total of four studies investigated the HRQOL [19,23,24,26]. One study reported significant improvement in physical functioning, less nausea and vomiting, and less pain with HB exercise than in the usual-care group [24]. The other two studies found a nonsignificant difference [23,26]. One study reported a 77.4% probability of inferiority (the Functional Assessment of Cancer Therapy–Prostate) of HB to CB exercise [19].

3. Future Directions

For further research on HB exercise intervention evaluations, we recommend researchers to:

• define the outcomes and evaluate the effect relevant to the phase of cancer rehabilitation,
• establish a timeframe to define the phase of cancer rehabilitation,
• determine relevant measurements of outcomes for different cancer populations, and
• conduct further pilot studies in understudied cancer populations to ensure the feasibility of interventions and data analysis for future research, e.g., sample size determination.

For the future methodological design of clinical trials with sufficient evidence of feasibility, we recommend using:

• specified eligibility criteria (clearly stated and fulfilled eligibility criteria),
• unbiased randomization with description (a description of the randomization method used to allocate patients into study groups should be provided),
• allocation concealment,
• blinding of the assessor (for at least one primary outcome), and
• intention-to-treat analysis.

For studies reporting HB exercise training interventions for cancer survivors during the rehabilitation phase, future studies are recommended to report: