Breast cancer was the most common cancer in women worldwide. The aims of the current systematic review and meta-analysis are: (i) to systematically examine the effects of exercise interventions on mental wellbeing; (ii) to examine the specific effect of the type of supervised exercise and its intensity, volume and frequency on mental wellbeing; and (iii) to explore which interventions are most effective in mental wellbeing among women with breast cancer during active treatment. An electronic literature search was performed using MEDLINE (via PubMed), Embase (Ovid), and Web of Science, we identified 175 full-text articles. The 57 publications included data from 6988 participants, age ranging from 18 to 78 years (weighted mean: 50.85 years). Compared with the control conditions, exercise training programs were associated with significant reductions in anxiety (d = −0.22, I2 = 53.0%), depression (d = −0.24, I2 = 66.6%), and fatigue (d = −0.47, I2 = 69.8%), as well as increases in body image (d = 0.27, I2 = 69.2%) and quality of life (overall, d = 0.46, I2 = 71.6%; emotional function, d = 0.33, I2 = 65.7%; and FACT-B, d = 0.60, I2 = 76.2%). There were a variety of frequencies, intensities, and durations of supervised exercise programs reported in the included meta-analytic approach. In addition, we found that concomitant concurrent training, at moderate-to-vigorous intensity, and with a volume ≥50 min/week had benefits on a number of health outcomes, such as fatigue, depression, and quality of life measure by the FACT-B instrument. These findings have important implications for healthcare providers and multidisciplinary teams involved in mental health management in cancer patients during active treatment.