The likelihood of being helped or harmed obtained from clinical trial results for cancer therapy: Can it really help?

Abstract

The majority of cancer clinical trials leading to therapeutic approval focus on outcomes such as objective tumor responses, progression-free survival (PFS), and overall survival (OS). However, it is equally important to assess toxicity, especially when comparing standard therapies with experimental ones. Clinical trials often fail to synthesize the relationship between efficacy and adverse event frequency, partly due to differences in measurement units. To address this, the number needed to treat (NNT) and number needed to harm (NNH) can be used as standardized measures. NNT represents the number of patients required to benefit from a treatment, while NNH indicates the number needed to experience harm. These metrics allow for a more balanced evaluation of therapeutic efficacy and toxicity. By calculating NNT for PFS or OS and NNH for adverse events, we can assess the therapeutic benefit relative to potential harm. The likelihood of being helped or harmed (LHH) combines these metrics into a ratio that expresses the balance between benefit and toxicity. Ideally, LHH values greater than 1 indicate a favorable balance toward efficacy. Though LHH has been applied mainly to psychotropic drugs, it was used in oncology sometimes. For example, studies in advanced non-small cell lung cancer and breast cancer have demonstrated LHH's utility in comparing treatments. Whereas LHH calculation has some limitations, it offers a valuable tool for explaining treatment risks and benefits to patients. It also could guide clinical trial design in cancer therapy.