#HEORInsights

Network Meta-Analysis (NMA) is a sophisticated statistical technique that allows researchers to simultaneously compare multiple treatments using a combination of direct and indirect evidence. While NMAs have revolutionized evidence synthesis, the inclusion of single-arm trials (SATs) within these analyses has sparked discussions and raised important questions about the potential challenges they bring.[1]

NMAs typically rely on a wealth of data, including randomized controlled trials (RCTs) and observational studies. However, there are instances where patient-level data from published studies might be inadequate, leading researchers to explore alternative avenues for incorporating evidence. This is where SATs come into play, offering a unique approach to addressing data gaps in NMA. SATs can function as mutual control groups based on the presence of complete covariate profiles and differing interventions. These trials primarily provide insights into the safety and efficacy of a treatment within a specific population. They often play a crucial role in early-phase drug development, helping researchers understand the potential benefits and risks of a novel intervention.[2-4]

The absence of a comparator group, which is the hallmark of SATs, is perhaps the greatest challenge of using these studies in NMAs because this precludes the construction of a connected network of treatments. NMAs depend on direct comparisons between treatments to establish relationships and estimate treatment effects accurately. The absence of head-to-head comparisons from SATs disrupts the seamless flow of evidence within the network. Moreover, the inclusion of SATs can potentially introduce bias and confounding. Without a control group, the observed outcomes in these trials might be influenced by factors beyond the intervention itself, leading to skewed estimates of treatment effects. This bias can jeopardize the validity and reliability of the overall NMA results.[5]

SATs can also exacerbate heterogeneity, which is an inherent challenge in evidence synthesis through NMAs. The diversity in study design, patient populations, and outcomes measured across these trials can contribute to significant heterogeneity within the NMA, undermining the assumption of treatment effect consistency that NMAs rely upon. The precision and confidence in NMA results can also be compromised when single-arm trials with limited sample sizes are included. Since these trials contribute a smaller amount of data relative to comparative trials, their impact on the overall estimates can be disproportionate. This can lead to wider confidence intervals and reduced statistical power, which in turn affects the reliability of the conclusions drawn from the NMA.[5]

To address these challenges, several strategies can be employed when considering the inclusion of SATs in NMAs. The incorporation of external evidence, such as historical controls, real-world data, or findings from related studies, can enrich the analysis by providing additional context and comparators, thereby enhancing the understanding of treatment efficacy and safety. Secondly, indirect comparisons can be leveraged, wherein treatments are linked through a common comparator, enabling the assessment of relative effects even in the absence of direct trial comparisons. Furthermore, the inclusion of matching controls entails identifying studies with analogous participant characteristics or baseline factors to the SAT, effectively generating a pseudo-comparator group that aligns with the trial’s attributes, thus mitigating potential bias. Moreover, the development of synthetic control groups by amalgamating data from multiple SATs that share comparable patient profiles and outcomes can be facilitated through statistical methods like propensity score matching or Bayesian approaches, enhancing the robustness of the analysis.[6-8]

Thus, SATs can offer valuable information for conducting NMAs, but their inclusion in the NMA is associated with some genuine challenges. By implementing strategies such as sensitivity analyses, cautious interpretation, and judicious inclusion, researchers can navigate these challenges and make well-informed decisions about including SATs in NMAs.

Become A Certified HEOR Professional – Enrol yourself here!

Learn AI Potential For Writing Manuscripts – Enrol yourself here!

References

1. Cipriani A, Higgins JP, Geddes JR, Salanti G. Conceptual and technical challenges in network meta-analysis. Annals of internal medicine. 2013 Jul 16;159(2):130-7.
2. Evans SR. Clinical trial structures. J Exp Stroke Transl Med. 2010 Feb 9;3(1):8-18.
3. Wang Z, Lin L, Murray T, Hodges JS, Chu H. Bridging randomized controlled trials and single-arm trials using commensurate priors in arm-based network meta-analysis. Ann Appl Stat. 2021 Dec;15(4):1767-1787.
4. Schmitz S, Maguire Á, Morris J, et al. The use of single armed observational data to closing the gap in otherwise disconnected evidence networks: a network meta-analysis in multiple myeloma. BMC Med Res Methodol. 2018 Jun 28;18(1):66.
5. Sachdev M. Using Single Arm Trials for HTA Submissions: Dr. Heeg and Maria Rizzo.
6. Seeger JD, Davis KJ, Iannacone MR, et al. Methods for external control groups for single arm trials or long-term uncontrolled extensions to randomized clinical trials. Pharmacoepidemiol Drug Saf. 2020 Nov;29(11):1382-1392.
7. Burcu M, Dreyer NA, Franklin JM, et al. Real-world evidence to support regulatory decision-making for medicines: Considerations for external control arms. Pharmacoepidemiol Drug Saf. 2020 Oct;29(10):1228-1235.
8. Zhang J, Ko CW, Nie L, Chen Y, Tiwari R. Bayesian hierarchical methods for meta-analysis combining randomized-controlled and single-arm studies. Statistical methods in medical research. 2019 May;28(5):1293-310.