Clinical trials offer patients with multiple myeloma (MM) treatment options that would otherwise be unavailable, and determine the future of MM approved drug regimens. However, there is a gap between the efficacy of a treatment in a clinical trial setting and the effectiveness it provides in real-world clinical practice. Paul G. Richardson from the Dana-Farber Cancer Institute, Boston, US, and collaborators, examined the possible reasons behind this apparent discrepancy focusing on patients with relapsed/refractory (R/R) MM, and presented their findings in Blood Cancer Journal in November 2018.
- Eligibility criteria for taking part in clinical trials vary considerably; trials recruit heterogeneous cohorts of patients and this may explain why there are noteworthy differences among endpoint values of treatment arms using the same regimen combinations
- Patient heterogeneity plays a major role in the differences observed between clinical trial and real-world outcomes. Data from the CONNECT-MM registry showed that up to 40% of patients with MM would not meet the stringent criteria that would allow them to enroll in a randomized clinical trial; trial-ineligible patients had a significantly lower three-year survival rate (63%) compared to trial-eligible patients (70%, P = 0.0392)
Factors contributing to patient heterogeneity include:
- Geographic variations: These may account for differences in populations, treatment options, and treatments that patients have been exposed to, before or after a clinical trial. Thus, clinical trial results taken from a specific geographic area may not translate into similar real-world outcomes for a population of another region
- Age and comorbidities: In MM, patient overall survival (OS) decreases with age and extensive comorbidities negatively affect response to treatment
- Disease stage: Although the International Staging System (ISS) stage is taken into account in MM trials, the number of patients with different ISS stages (I, II, or III) varies throughout different clinical trials and also between the clinical trial and real-world setting. Despite the fact that the revised ISS (R-ISS) has stronger prognostic value, it is still not widely used in the clinical trial environment or the community clinical practice
- Disease subtype: MM prognosis depends on the disease type (non-secretory, oligo-secretory, secretory) and amounts of free light chain (levels and ratio); these values are not always considered, but can affect clinical outcomes differently
- Renal impairment: Despite the fact that around 59% of R/R MM patients have a history of renal impairment, this is a common exclusion criterion in MM clinical trials, although the cut-off value of creatinine clearance is not standardized
- Cytogenetic abnormalities (CAs): It is well recognized that certain CAs define a group of patients with high-risk MM. However, CAs are not always recorded in clinical trials and the patient’s CA profile is not routinely examined in real-world settings. In addition, it is still not clear how the relationship between CAs and MM clone size impact prognosis
- Prior treatment exposure: Number and types of previous treatments, and refractoriness to previous regimens affects response to salvage therapy
- Clinical trial design: Interpretation of clinical trial results can be impacted in the case of a non-blinded clinical trial
- Endpoint data interpretation: Different published criteria define MM clinical trial endpoints differently and translating these endpoints to real-world practice is not always possible.
- Effects of toxicity and aggressive relapse may affect response rates and association with OS
- Minimal residual disease (MRD) also contributes to outcome interpretation; many MM clinical trials are now incorporating measurement of MRD but this is still not standard practice and is not performed in a homogeneous way, impeding direct comparison of results between different trials. Furthermore, MRD is not measured routinely in the real-world setting
- Patient-reported outcomes are determined by the timing and method used to collect and analyze them
Direct comparisons across clinical trials and between clinical trials and real-world outcomes are challenging. A wide range of factors inhibits immediate translation of the efficacy of a drug regimen from the clinical trial environment to an effective treatment in clinical practice. A way around this issue is by using indirect comparisons of relative efficacy versus a common comparator with the implementation of hazard ratios.