Rare Disease Day 2026: From Diagnosis to the Operating Room — How AI and Genetic Testing Are Transforming Precision Care for Rare Skeletal Disorders

Rare diseases collectively affect more than 400 million individuals worldwide, and genetic alterations account for approximately 80% of cases. Today, over 500 genes have been linked to recognized genetic skeletal disorders, and this number continues to rise with advances in sequencing technology and improvements in data interpretation. Despite this genomic revolution, patients with rare skeletal disorders still face prolonged diagnostic journeys. Delays in diagnosis can have serious consequences, such as delayed treatment and inappropriate intervention.

With Rare Disease Day 2026 just around the corner, it is important to highlight two developments that are helping to close this gap. First, artificial intelligence (AI) is enabling faster and more accurate diagnoses. Second, genetic testing is guiding surgical decisions, allowing precision spine care.

Deep Phenotyping and AI: Unlocking the Diagnostic Puzzle

One of the most frustrating realities in rare skeletal medicine is that expanded genomic sequencing has not automatically shortened the time to diagnosis. The challenge is not a lack of genetic data but a shortage of the right clinical data to interpret it. Rare skeletal disorders often present with overlapping phenotypes, variable expressivity, and age-dependent manifestations. Patients with the same genetic variant may look very different at different stages of life. Besides, subtle features — vertebral segmentation patterns, limb proportions, epiphyseal development, joint hypermobility, and extra-skeletal findings — often carry crucial diagnostic information but are easily overlooked in standard clinical assessments.

These challenges can be addressed through deep phenotyping, a systematic approach to capturing clinical features across organ systems and over time. Systematic deep phenotyping through the nine strategic categories is pivotal for closing the “sequencing-analysis” gap in rare disease diagnostics. The iDREAMS framework provides a practical and scalable model that enhances diagnostic accuracy, lowers the entry barrier for resource-limited settings, and extends the reach of precision medicine to patients who had long remained undiagnosed. In addition, tools such as the Human Phenotype Ontology support standardization across centers and improve reproducibility. Collectively, these strategies could elucidate disease heterogeneity and strengthen genotype–phenotype correlations. They enable phenotype-first variant interpretation and iterative reanalysis, while facilitating the discovery of novel biomarkers. These advances improve diagnostic accuracy, refine disease stratification, and guide personalized care. They also support biomarker-driven clinical trials and contribute to disease prevention, more precise prognostic prediction, and lifespan-oriented management strategies spanning pregnancy to aging.

Further, AI amplifies this approach considerably. Natural language processing can extract structured phenotypic data from unstructured clinical records, which is otherwise enormously time-consuming. Image-based AI models assist in analyzing radiographs and advanced imaging for subtle morphological abnormalities that may escape the human eye. When integrated with genomic decision-support platforms, AI can prioritize candidate variants and integrate complex phenotypic profiles, boosting diagnostic yield in ways that were simply not possible a decade ago. Importantly, AI complements rather than replaces human expertise. A human-in-the-loop approach ensures that AI recommendations are interpreted by experienced clinicians who can contextualize results, account for nuance, and bear responsibility for patient care. AI should be understood as a powerful tool in expert hands.

From Diagnosis to the Operating Room: Genetics-Informed Spine Surgery

Accurate genetic diagnosis is not an endpoint. It is the starting point for clinical decision making, which is increasingly extending into the operating room in spine surgery. Spinal deformity is a common feature of many rare genetic disorders. These conditions include Ehlers-Danlos syndrome, Marfan syndrome, Neurofibromatosis type 1, and Down syndrome. Deformities often develop early in life and may progress rapidly. When left untreated, they can lead to chronic pain, neurologic deficits, cardiopulmonary compromise, and in some cases reduced life expectancy. Nonoperative management is frequently less successful in these patients, often leaving surgical correction as the only option. However, surgical intervention in patients with genetic syndromes carries a higher risk than in the general population. Many of these risks are directly related to the underlying genetic diagnosis. For example, patients with Ehlers-Danlos syndrome have fragile connective tissue, which increases the risk of poor wound healing, excessive bleeding, and failure of surgical fixation.

To address this gap, recent studies have begun to evaluate the use of preoperative genetic testing to inform surgical strategy, timing, and risk reduction. Early evidence suggests that genetic diagnosis can meaningfully influence surgical outcomes. For example, patients with TBX6-associated congenital scoliosis demonstrate more favorable surgical outcomes compared with patients who have other causes of congenital scoliosis. These findings support the role of genetically informed surgical planning. In hereditary connective tissue disorders, including Ehlers-Danlos syndrome, Marfan syndrome, and Loeys-Dietz syndrome, genetic testing can help identify the risk of postoperative complications that may not be clinically apparent before surgery. One such complication is the adding-on phenomenon, which is thought to be driven by underlying spinal hypermobility. Recognizing this risk preoperatively can support the decision to perform a longer fusion to reduce the likelihood of revision surgery. In addition, altered vertebral anatomy and reduced bone mineral density have been recognized in these conditions and may contribute to higher rates of implant-related complications. These findings highlight the need for further study to guide the selection of more appropriate fixation systems and surgical techniques. Another compelling example of how genetic diagnosis can shape treatment regimens is spinal muscular atrophy. Disease-modifying therapies, such as nusinersen, risdiplam, and gene therapy, have changed the natural history of scoliosis. For patients with Type II and Type III SMA, who previously faced a substantial disease burden, these therapies can slow scoliosis progression and improve motor function, which may change the optimal timing for surgery. In contrast, patients with Type I SMA, who gain the ability to sit or stand after treatment, are now at risk of developing scoliosis. This new possibility requires earlier monitoring and careful consideration of surgical planning.

A Unified Vision: Toward Precision Care for Rare Bone Diseases

Surgery has long been viewed as a “technician’s science,” but a clear clinical rationale in both diagnosis and treatment is essential for shared decision-making and optimal patient outcomes. The integration of AI, deep phenotyping, and genetics represents a new paradigm in rare disease management. Earlier and more accurate diagnoses reduce misdiagnosis and allow for timely intervention. Genetics-informed surgical planning ensures that interventions are tailored to individual risk profiles, anatomical variations, and evolving therapies.

Rare Disease Day 2026 continues to drive global collaboration across genetics, physicians, and scientists. By bringing together multidisciplinary expertise and cutting-edge techniques, we can accelerate scientific discovery and translate it into patient-centered care in the era of precision medicine. Together, we are bringing hope and a brighter future to patients and families around the world.

References

1. Kernohan KD, Boycott KM: The expanding diagnostic toolbox for rare genetic diseases. Nat Rev Genet 2024, 25(6):401-415.
2. Robinson PN: Deep phenotyping for precision medicine. Hum Mutat 2012, 33(5):777-780.
3. Li G, Xu K, Cai J, Jiang Y, Zhou X, Lv Y, Hao N, Yu Y, Zhao S, Li Q et al: Deep insights and clinical benefits from the comprehensive cohort of fetal skeletal dysplasia in China. J Genet Genomics 2025.
4. Unger S, Ferreira CR, Mortier GR, Ali H, Bertola DR, Calder A, Cohn DH, Cormier-Daire V, Girisha KM, Hall C et al: Nosology of genetic skeletal disorders: 2023 revision. Am J Med Genet A 2023, 191(5):1164-1209.
5. Li G, Xu K, Yin X, Yang J, Cai J, Yang X, Li Q, Wang J, Zhao Z, Mahesahti A et al: Integrating deep phenotyping with genetic analysis: a comprehensive workflow for diagnosis and management of rare bone diseases. Orphanet J Rare Dis 2024, 19(1):371.
6. Kanjee Z, Crowe B, Rodman A: Accuracy of a Generative Artificial Intelligence Model in a Complex Diagnostic Challenge. JAMA 2023, 330(1):78-80.
7. Javanmardi B, Waikel RL, Tkemaladze T, Moosa S, Küsshauer A, Pantel JT, Fardipour M, Krawitz P, Solomon BD, Mohnike K: Artificial intelligence for diagnosing rare bone diseases: a global survey of healthcare professionals. Orphanet J Rare Dis 2025, 20(1):365.
8. Li, J.G., K. Xu, B. Xiao, J. Li, Y.C. Zhu, H. Jin, et al., Addressing the diagnostic gap through deep phenotyping. Hum Genomics, 2026.
9. Luo L, Yan S, Lai PT, Veltri D, Oler A, Xirasagar S, Ghosh R, Similuk M, Robinson PN, Lu Z: PhenoTagger: a hybrid method for phenotype concept recognition using human phenotype ontology. Bioinformatics 2021, 37(13):1884-1890.
10. Feng Y, Qi L, Tian W: PhenoBERT: A Combined Deep Learning Method for Automated Recognition of Human Phenotype Ontology. IEEE/ACM Trans Comput Biol Bioinform 2023, 20(2):1269-1277.
11. Yang J, Liu C, Deng W, Wu D, Weng C, Zhou Y, Wang K: Enhancing phenotype recognition in clinical notes using large language models: PhenoBCBERT and PhenoGPT. Patterns (N Y) 2024, 5(1):100887.
12. Gargano MA, Matentzoglu N, Coleman B, Addo-Lartey EB, Anagnostopoulos AV, Anderton J, Avillach P, Bagley AM, Bakštein E, Balhoff JP et al: The Human Phenotype Ontology in 2024: phenotypes around the world. Nucleic Acids Res 2024, 52(D1):D1333-d1346.
13. Jaganathan K, Kyriazopoulou Panagiotopoulou S, McRae JF, Darbandi SF, Knowles D, Li YI, Kosmicki JA, Arbelaez J, Cui W, Schwartz GB et al: Predicting Splicing from Primary Sequence with Deep Learning. Cell 2019, 176(3):535-548.e524.
14. Ioannidis NM, Rothstein JH, Pejaver V, Middha S, McDonnell SK, Baheti S, Musolf A, Li Q, Holzinger E, Karyadi D et al: REVEL: An Ensemble Method for Predicting the Pathogenicity of Rare Missense Variants. Am J Hum Genet 2016, 99(4):877-885.
15. Cheng J, Novati G, Pan J, Bycroft C, Žemgulytė A, Applebaum T, Pritzel A, Wong LH, Zielinski M, Sargeant T et al: Accurate proteome-wide missense variant effect prediction with AlphaMissense. Science 2023, 381(6664):eadg7492.
16. Son JH, Xie G, Yuan C, Ena L, Li Z, Goldstein A, Huang L, Wang L, Shen F, Liu H et al: Deep Phenotyping on Electronic Health Records Facilitates Genetic Diagnosis by Clinical Exomes. Am J Hum Genet 2018, 103(1):58-73.
17. Chen Z, Zheng Y, Yang Y, Huang Y, Zhao S, Zhao H, Yu C, Dong X, Zhang Y, Wang L et al: PhenoApt leverages clinical expertise to prioritize candidate genes via machine learning. Am J Hum Genet 2022, 109(2):270-281.
18. Smedley D, Jacobsen JO, Jäger M, Köhler S, Holtgrewe M, Schubach M, Siragusa E, Zemojtel T, Buske OJ, Washington NL et al: Next-generation diagnostics and disease-gene discovery with the Exomiser. Nat Protoc 2015, 10(12):2004-2015.
19. Barile J, Margolis A, Cason G, Kim R, Kalash S, Tchaconas A, Milanaik R: Diagnostic Accuracy of a Large Language Model in Pediatric Case Studies. JAMA Pediatr 2024, 178(3):313-315.
20. Yurkovich JT, Evans SJ, Rappaport N, Boore JL, Lovejoy JC, Price ND, Hood LE: The transition from genomics to phenomics in personalized population health. Nat Rev Genet 2024, 25(4):286-302.
21. Xu K, Li G, Wu Z, Zhang TJ, Wu N: Diagnosis and treatment of the Ehlers-Danlos syndromes in China: synopsis of the first guidelines. Orphanet J Rare Dis 2024, 19(1):194.
22. Garcia BT, Westerfield L, Yelemali P, Gogate N, Andres Rivera-Munoz E, Du H, Dawood M, Jolly A, Lupski JR, Posey JE: Improving Automated Deep Phenotyping Through Large Language Models Using Retrieval Augmented Generation. medRxiv 2024.
23. Tesi B, Boileau C, Boycott KM, Canaud G, Caulfield M, Choukair D, Hill S, Spielmann M, Wedell A, Wirta V et al: Precision medicine in rare diseases: What is next? J Intern Med 2023, 294(4):397-412.
24. Rohrbach M, Giunta C: PLOD1-Related Kyphoscoliotic Ehlers-Danlos Syndrome. In: GeneReviews((R)). edn. Edited by Adam MP, Bick S, Mirzaa GM, Pagon RA, Wallace SE, Amemiya A. Seattle (WA); 1993.
25. Welborn M, Tambe A, Adeyemi A, Dupuis M, Simoneau D, Brandi ML: Burden of disease and unmet needs associated with scoliosis in neurofibromatosis type 1: a systematic literature review. JBMR Plus 2025, 9(8):ziaf072.
26. Rako K, Ranade S, Allen A: Orthopaedic Management in Down Syndrome. Journal of the Pediatric Orthopaedic Society of North America 2021, 3(2):283.
27. Pehrsson K, Larsson S, Oden A, Nachemson A: Long-term follow-up of patients with untreated scoliosis. A study of mortality, causes of death, and symptoms. Spine (Phila Pa 1976) 1992, 17(9):1091-1096.
28. Hawes MC: Health and function of patients with untreated idiopathic scoliosis. JAMA 2003, 289(20):2644; author reply 2644-2645.
29. Levy BJ, Schulz JF, Fornari ED, Wollowick AL: Complications associated with surgical repair of syndromic scoliosis. Scoliosis 2015, 10:14.
30. Yonko EA, LoTurco HM, Carter EM, Raggio CL: Orthopedic considerations and surgical outcomes in Ehlers-Danlos syndromes. Am J Med Genet C Semin Med Genet 2021, 187(4):458-465.
31. Wu N, Ming X, Xiao J, Wu Z, Chen X, Shinawi M, Shen Y, Yu G, Liu J, Xie H et al: TBX6 null variants and a common hypomorphic allele in congenital scoliosis. N Engl J Med 2015, 372(4):341-350.
32. Lin G, Yang Y, Chen Z, Zhao S, Niu Y, Du Y, Zhao Y, Wang S, Wu N, Zhang J: Preliminary study assessing the long-term surgical outcomes of TBX6-associated congenital scoliosis (TACS) patients using the propensity score matching method: exploring the clinical implications of genetic discoveries in congenital scoliosis. Orphanet J Rare Dis 2025, 20(1):32.
33. Bridwell KH, Gupta M: Bridwell and DeWald's textbook of spinal surgery: Lippincott Williams & Wilkins; 2019.
34. DeWald C, DeWald R: Spinal Deformities: The Comprehensive Text: Thieme Medical Publishers; 2024.
35. Maheshati A, Xu K, Li Z, Li G, Yin X, Li Q, Liu D, Wang S, Wu Z, Qiu G et al: Selecting the Substantially Touched Vertebra as the Lowest Instrumented Vertebra in Spinal Surgeries for B3GALT6 -Related Disorders: Clinical Experience and Literature Review. Orthop Surg 2025, 17(7):2025-2037.
36. Gnazzo M, Pisano G, Piccolo B, Turco EC, Esposito S, Pera MC: Scoliosis in spinal muscular atrophy in the era of disease-modifying therapy: a scoping review. Neurol Sci 2025, 46(8):3431-3442.
37. Ip HNH, Yu MKL, Wong WHS, Liu A, Kwan KYH, Chan SHS: Treatment of Symptomatic Spinal Muscular Atrophy with Nusinersen: A Prospective Longitudinal Study on Scoliosis Progression. J Neuromuscul Dis 2024, 11(2):349-359.