Objective: A strong therapeutic alliance has been found to predict psychotherapeutic treatment success across a variety of therapeutic modalities and patient populations. However, only a few studies have examined therapeutic alliance as a predictor of psychotherapy outcome among cancer survivors, and none have examined this relation in telephone-administered cognitive-behavioral therapy (T-CBT). This study evaluated the extent to which therapeutic alliance affected psychotherapy outcomes in survivors of hematopoietic stem cell transplantation (HSCT), a treatment for some cancers. Method: Forty-six patients enrolled in a randomized clinical trial of T-CBT for posttraumatic stress disorder (PTSD) completed a baseline assessment (including self-report measures of PTSD symptoms, depression, and general distress), 10 individual T-CBT sessions, and follow-up assessments at 6, 9, and 12 months post-baseline. Therapeutic alliance was assessed after the 3rd T-CBT session with the Working Alliance Inventory, which yields overall and subscale (task, bond, and goal) scores. Results: Analyses revealed that higher total therapeutic alliance scores prospectively predicted decreased depressive symptomatology; higher task scores predicted decreased overall distress, depressive symptomatology, symptoms of re-experiencing, and avoidance; and higher bond scores predicted decreased depressive symptomatology and symptoms of re-experiencing. Conclusions: These results suggest that assessments of therapeutic alliance should be incorporated into routine clinical care, and therapeutic alliance should be specifically cultivated in interventions to maximize psychotherapeutic benefits involving vulnerable populations such as cancer survivors. (Contains 4 tables.)