1. Home-based aerobic training: The aerobic exercise consisted of a walking program based on the number of steps per day, which was recorded using a pedometer (Yamax Digiwalker SW-200, Tokyo, Japan). The target of the aerobic exercise was individually tailored by adding 3000 accumulated steps above their baseline value on 5 days of the week.
2. Home-based resistance training: The resistance training was conducted using a set of resistance bands (Happy Fit®, Jakarta, Indonesia). This training consisted of two sets of biceps curl, triceps extension, chest press, shoulder flexion, shoulder press, lateral extension, seated row, and upright row, with 8-10 repetitions. Patients were instructed to perform each exercise with an average total duration of 10 minutes on 2 days a week at home. The exercises were individually tailored based on participants' strength and range of movement using a set of three color-coded bands representing low, moderate, and high levels of resistance.

Physical fitness was measured using a submaximal treadmill test with the modified Bruce protocol. Treadmill speed and inclination were automatically increased every 3 min (test stage). The test was terminated when participants reached volitional fatigue. Physical fitness was classified as low, fair, average, and good, based on patients' age, treadmill test duration, and predicted VO2peak.

Fatigue was assessed using the Indonesian version of Fatigue Severity Scale (FSS). A higher score on the FSS indicated a more severe level of fatigue. The FSS was also classified as no-to-mild fatigue (FSS <4.0) and severe fatigue (≥4.0).

Health-related quality of life was assessed using the Indonesian version of the EORTC Quality of Life Questionnaire (QLQ-C30). The EORTC QLQ-C30 consists of global health status, functional scales, and symptom scales. A higher score indicates better health for global health status and functional scales, but represents a higher level of burden for symptom scales.

Venous blood (5 ml) was collected at baseline and post-intervention to measure the levels of inflammatory biomarkers, including CRP, IL-6, and IL-10. The CRP levels were measured using an Enzyme-Linked Immunosorbent Assay (ELISA) assay (Calbiotech, CR120C), while IL-6 and IL-10 were measured using a Luminex assay (eBioscience, PPX-02-MXYMMK2).

Intervention feasibility was determined by (a) recruitment (ratio of the number of participants recruited to the number of eligible participants approached), (b) retention (ratio of the number of participants who completed the 12-week program to the total number of participants enrolled in the study), (c) adherence (the ratio of the number of participants meeting at least 70% of prescribed activities calculated using steps per day, resistance exercise frequency, duration, and intensity to the total number of participants completing the 12-week program). The intervention was defined as feasible if those rates were above 75%.