Accuracy of the molecular diagnosis of Duchenne and Becker muscular dystrophy: A systematic review with meta-analysis
Clinical Snapshot
PICO Framework
| P — Population | Patients with clinical suspicion of Duchenne muscular dystrophy (DMD) or Becker muscular dystrophy (BMD), across all age groups and sexes |
| I — Intervention | Multiplex ligation-dependent probe amplification (MLPA), next-generation sequencing (NGS), or the sequential algorithm MLPA followed by NGS (MLPA-NGS) |
| C — Comparator | Multiplex PCR or muscle biopsy as the reference standard for confirmatory diagnosis |
| O — Outcomes | Diagnostic accuracy metrics: sensitivity, specificity, area under the ROC curve (AUC), and detection rate for each molecular diagnostic strategy |
Bottom Line
This systematic review and meta-analysis of 24 studies (over 8,000 patients) provides the most comprehensive pooled evidence to date on molecular diagnostic accuracy for DMD and BMD. MLPA demonstrates good overall performance (sensitivity 80%, specificity 93%, AUC 0.90) and should remain the first-line molecular test given its accessibility and cost-effectiveness. Critically, a sequential strategy — performing NGS in MLPA-negative patients who retain clinical suspicion — achieves a detection rate of 97%, representing a clinically meaningful improvement over either test alone. This stepwise approach is operationally rational and cost-efficient. However, clinicians must interpret these pooled figures with caution: heterogeneity is very high across all estimates (I²=86–95%), and GRADE certainty is rated as very low. The wide confidence interval for NGS alone (61–87%) reflects substantial variability across study populations, likely driven by differences in mutation spectrum, reference standards, and patient selection. Prediction intervals — which would better reflect real-world variability — are not reported. In Australian practice, the sequential MLPA-NGS algorithm aligns with existing clinical genetics pathways and is directly relevant to mutation-specific therapy eligibility. Definitive guideline-level recommendations await higher-certainty evidence from prospectively designed diagnostic accuracy studies with standardised reference standards.
Key Findings
P Value: Not reported in abstract
Effect Size: MLPA: sensitivity 0.80, specificity 0.93, AUC 0.90; NGS alone: detection rate 0.77; Sequential MLPA-NGS: detection rate 0.97
Primary Outcome: Diagnostic accuracy of MLPA, NGS, and sequential MLPA-NGS for confirmatory diagnosis of DMD/BMD in patients with clinical suspicion
Nnt Or Sensitivity: MLPA sensitivity 80% (95% CI: 76–84%); MLPA specificity 93% (95% CI: 87–96%); Sequential MLPA-NGS detection rate 97% (95% CI: 94–99%). Note: very high heterogeneity (I²=86–95%) across all estimates substantially limits clinical applicability of these pooled figures
Confidence Interval: MLPA sensitivity 95% CI: 0.76–0.84; MLPA specificity 95% CI: 0.87–0.96; MLPA AUC 95% CI: 0.89–0.92; NGS detection rate 95% CI: 0.61–0.87; MLPA-NGS detection rate 95% CI: 0.94–0.99
Clinical Application
MLPA is widely available in accredited molecular genetics laboratories, relatively cost-effective, and technically straightforward. NGS (whole-exome or targeted dystrophin gene panels) is increasingly accessible but remains more expensive and requires greater bioinformatic infrastructure. The sequential algorithm (MLPA first, NGS for MLPA-negative cases) is operationally feasible and cost-efficient, as NGS is reserved for the subset of patients not diagnosed by MLPA. In Australia, molecular diagnosis of DMD/BMD is available through state-based clinical genetics services and the National Pathology Accreditation Advisory Council (NPAAC)-accredited laboratories. MLPA for dystrophin gene deletions and duplications is Medicare-rebatable under specific item numbers for patients with clinical suspicion of DMD/BMD. NGS-based dystrophin gene sequencing is increasingly available through paediatric neurology and clinical genetics services, including through the Australian Genomics Health Alliance framework. The sequential MLPA-NGS approach aligns with current RACGP and ACMG guidance on stepwise genetic testing. Exon-skipping therapies (e.g., eteplirsen, golodirsen) require precise molecular characterisation of the causative mutation for eligibility determination, making accurate molecular diagnosis directly relevant to access of emerging TGA-approved or compassionate access therapies. Genetic counselling services are integral to the diagnostic pathway in Australia, and a confirmed molecular diagnosis is essential for cascade family testing and prenatal diagnosis. Male patients (and female carriers) with clinical suspicion of DMD or BMD based on elevated creatine kinase, characteristic clinical phenotype, or family history. Most applicable to paediatric neurology and clinical genetics settings. Results are most directly applicable to populations with a mutation spectrum similar to those in the included studies.
Abstract
INTRODUCTION: Recently, Molecular diagnosis of Duchenne muscular dystrophy (DMD) and Becker muscular dystrophy (BMD) has become increasingly important in the management of these patients, with techniques such as multiplex ligation-dependent amplification (MLPA) and next-generation sequencing (NGS) coming to the fore. Therefore, this study aims to evaluate the diagnostic accuracy of MLPA, NGS, and the algorithm MLPA-NGS for confirmatory diagnosis of DMD/BMD. METHODS: We systematically searched databases (PubMed, Embase, Scopus, Cochrane and Web of Science) until July 2025 for studies evaluating the diagnostic accuracy of MLPA and/or NGS testing in patients with clinical suspicion of DMD, considering multiplex PCR or biopsy as the reference test. A meta-analysis was performed using a random-effects model to estimate the sensitivity, specificity, and detection rate of each test. The QUADAS-2 tool was used to assess the risk of bias and the GRADE criteria were used to identify the certainty of evidence. RESULTS: We included 10 studies (3786 patients) evaluating the use of MLPA and 14 studies (4333 patients) evaluating the use of NGS. For MLPA, the sensitivity was 0.80 (95%CI: 0.76-0.84; I2: 86%), the specificity was 0.93 (95%CI: 0.87-0.96; I2: 16%), and AUC 0.90 (CI-95%: 0.89-0.92). For NGS, the detection rate was 0.77 (95% CI: 0.61-0.87; I2: 94%). Furthermore, the detection rate increased to 0.97 (95% CI: 0.94-0.99; I2: 95%) when NGS was performed after MLPA. We observe a low risk of bias but with very low certainty in the estimations. CONCLUSIONS: In patients with clinical suspicion of DMD, the MLPA test is very good but with very low certainty. However, in these patients with a negative MLPA, adding the NGS test would allow improve the detection rate. Therefore, the sequential use of these tests could be considered in patients who persist in the clinical suspicion.
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service