Part 1 of a 2-part introduction to the neural origin of scoliosis. Part 2: Not Everyone Has the Gene. Here’s What’s Actually Generating the Curve. For the full synthesis, see The Scoliosis Research That Points to the Nervous System.
This is one of the most searched questions about scoliosis, and the answer is yes, partly. A Swedish study of more than 64,000 twins put the heritability of scoliosis at about 38%. There are genes involved.
But look at what those genes do. Some shape the spine itself: PAX1 and GPR126 act on growth plates, cartilage and discs. One of the strongest signals sits near LBX1, a gene that helps build sensory nerve cells in the spinal cord. And a rarer gene, POC5, helps build cilia, tiny hair-like structures found on many cells, including in the inner ear.
So the genes point at the structure, and at least one points toward the sensing.
That second direction is the one most people never hear about, and it is what this article is about.
None of these genes decides a curve on its own. The effects are small, and most people who carry the risk variants never develop scoliosis.
Is scoliosis genetic?
Partly. A Swedish twin-registry study of 64,578 twins estimated the heritability of self-reported scoliosis of any type at about 38% (Grauers et al., 2012). Among replicated risk loci for adolescent idiopathic scoliosis, PAX1 and GPR126 act on growth plate, cartilage and disc; another lies near LBX1, a gene that in mice specifies dorsal-horn sensory interneurons; whether risk at the LBX1 locus acts through those neurons is not confirmed. Rare variants in POC5, a centriole and cilia gene, have been found in families with idiopathic scoliosis (Patten et al., 2015). Effects are small and most carriers never develop a curve.
LBX1: one of the strongest genetic signals
LBX1 sits at one of the strongest and best-replicated genetic signals for adolescent idiopathic scoliosis, found across several populations.
In mice, LBX1 builds one of the two main classes of sensory interneurons in the spinal cord: cells that handle touch and pain, some of them carrying signals about where the body is. It does not build bone.
That matters because of what those signals feed. Your brain keeps a running map of your body, and it uses position signals from the trunk to keep that map accurate. If those signals were blurrier, the map would be working with worse data.
Whether the human risk at LBX1 actually works that way is not yet confirmed. The same stretch of DNA also affects muscle and body-axis development. But one of the strongest signals for a spinal curve sitting near a gene that helps build sensory nerve cells is hard to ignore.
What does LBX1 have to do with scoliosis?
The LBX1 locus is one of the strongest and most replicated genetic associations with adolescent idiopathic scoliosis (Takahashi et al., 2011; meta-analysis of 34,626 subjects, Cao et al., 2016; pooled odds ratio about 1.2). In mice, Lbx1 is required for one of the two major classes of dorsal-horn sensory interneurons, which relay pain and touch, some of which handle position signals in the spinal cord (Gross et al., 2002; Satoh et al., 2016). Lbx1 does not encode a bone or cartilage protein, but the locus also acts in muscle and body-axis patterning, and whether AIS risk there acts through sensory neurons is not confirmed.
POC5: a gene for cilia
POC5 is a rarer finding. In 2015, researchers traced scoliosis through a large family and found a variant in POC5 that travelled with the condition. They then found POC5 variants in other families and patients.
POC5 is part of the machinery cells use to build centrioles and cilia. When the researchers put the human variants into zebrafish, the fish developed spine deformity without other skeletal defects. Later cell studies linked the variant to shorter cilia.
Cilia matter in many tissues, including the inner ear and bone-forming cells. In zebrafish, the best-studied cilia route to scoliosis runs through fluid flow in the spinal canal, not through sensing (Grimes et al., 2016). It is not yet known which tissue matters for POC5 in people. It is a lead, not an answer.
What the best-studied genes point at
More than 20 risk regions for scoliosis have been found, and most have not been traced to a tissue. A few of the best-studied ones have. Structural genes such as PAX1 and GPR126 act on cartilage, growth plates and discs. LBX1 sits next to a gene that builds sensory nerve cells, though the same DNA also affects muscle. POC5 is a cilia gene; in fish, the best-studied cilia route to scoliosis runs through fluid flow in the spinal canal, not sensing. CHD7 is a candidate with weaker evidence.
So scoliosis may not be only a spine problem.
It may also be a sensing problem that shows up in the spine. That is a big idea built on a handful of findings, and the paper treats it as one to test, not a conclusion.
The Neural Generation Hypothesis builds on that second half. It proposes that when the sense of where the trunk is gets blurry during the growth spurt, slow drift is too small to notice, and the brain’s map of the body updates toward the curve instead of back to straight. That would hold the part of the curve a person can actively correct. Growth and uneven loading still decide how big the curve gets.
What the chicken experiments show, and what they don’t
Remove the pineal gland from newly hatched chickens and many of them develop scoliosis. This is a hormonal experiment, not a sensory one. In one 1995 study, all 30 untreated chickens did. Giving them melatonin, the hormone the pineal gland makes, cut that to 6 of 30.
The authors suggested that losing melatonin disturbs the symmetrical growth of the position-sense system around the spine. That is their reading. They did not measure it.
The result also does not travel well.
In another study, 10 of 21 chickens curved after the surgery, while rats and hamsters did not. And in 18 young rhesus monkeys, much closer relatives of ours, 10 of them with the gland confirmed fully removed, the same surgery produced no scoliosis at all. Findings in chickens cannot simply be applied to people.
The frog studies: a clear animal result
In 2009, Lambert and colleagues removed the balance organ on one side in frog tadpoles. As adults, the frogs had spines curved sideways and front to back, with rotation and misshapen vertebrae, much like human scoliosis.
Why frogs? Land animals usually recover from losing one balance organ, because their limbs, pressing on the ground, supply backup position sense. Tadpoles swim, so there is no backup. The uneven signal to the trunk muscles persists. A 2013 follow-up traced the change in the nerve pathways and proposed that the constant uneven pull bends the soft, growing skeleton.
It is a clear result. It is also a frog. A one-sided injury in an amphibian produces a curve through uneven muscle pull, which is closer to scoliosis after nerve damage than to the ordinary teenage kind. What it shows is that the balance system matters for keeping a spine straight.
Can losing balance or position sense cause scoliosis in animals?
In some animals, yes. Unilateral removal of the labyrinth in Xenopus tadpoles produced adult spinal deformation with lateral and sagittal curvature and vertebral rotation (Lambert et al., 2009), attributed to persistent asymmetric vestibulospinal drive (Lambert et al., 2013). Pinealectomy, a hormonal rather than sensory manipulation, produced scoliosis in chickens (Machida et al., 1995; 10 of 21 in O’Kelly et al., 1999) but not in rats, hamsters (O’Kelly et al., 1999) or 18 rhesus monkeys (10 with confirmed complete removal; Cheung et al., 2005). One-sided lesion models test a different mechanism from ordinary adolescent idiopathic scoliosis, and none of these results shows that the same process operates in humans.
Your inner ear and your spine
The vestibular system in your inner ear does more than keep your balance. It tells your brain which way is up, and that signal runs down to the muscles of your trunk.
People with scoliosis do show differences here. In two small MRI studies, the shape of one inner-ear canal differed in teenagers with AIS, an anatomy the authors argue is probably set before birth. But the two studies disagree on which way the canal is tilted. A 2022 review of five such studies found asymmetries mainly on the left side, linked to which part of the spine is curved, but not to the curve’s side or size.
The Neural Generation Hypothesis adds one more proposal, not yet tested: that the inner ear’s sense of vertical is itself calibrated against position signals from the neck, so a blurry neck signal could tilt the reference too.
The weighting loop
Your brain blends signals from your eyes, your inner ear and your position sense, and it gives more weight to whichever it trusts most.
Model-based estimates from one research group suggest teenagers with AIS give inner-ear input about twice the weight controls do, roughly 13% against 6%. A 2025 study found their balance responses to inner-ear stimulation were larger than normal, but symmetric and pointing the normal way. Two Czech studies, from 2011 and 2025, found the visual sense of vertical slightly off (in 2025, only during motion), though other studies found it normal; the 2025 study also found abnormal inner-ear reflex tests in 38% of 20 patients. A Chinese study the same year found more sway on soft surfaces.
One reading: the brain leans harder on a channel that may itself be off, and that keeps the pattern going. Another: these are adjustments to living in a curved body. All of it was measured after the curve appeared. The studies cannot yet tell which.
Is the vestibular system different in adolescent idiopathic scoliosis?
Several studies report vestibular differences in adolescent idiopathic scoliosis (AIS), all measured after onset. Using a neuromechanical model, Pialasse et al. (2015) estimated about 13% vestibular weighting in AIS versus about 6% in controls. Cyr et al. (2025) found larger vestibular-evoked balance responses in AIS that were symmetric and normally oriented. Cakrt et al. (2011) and Kučerová et al. (2025, in dynamic, not static, testing) reported altered subjective visual vertical, and Kučerová et al. found abnormal VEMP responses in 38% of patients (of 20). MRI studies report lateral semicircular canal asymmetry (Hitier et al., 2015; Carry et al., 2020), though they disagree on direction. Whether these differences contribute to the curve or follow from it is not established.
Why this is actually good news
If scoliosis were purely a bone problem, the options would be mostly mechanical: bracing, surgery and scoliosis-specific exercise.
But sensing can be trained. Balance, position sense and the way the brain weights its signals all respond to practice. Whether training them changes a curve is still an open question. One 2025 trial in 62 teenagers with mild curves found that swapping part of a Schroth session for sensory integration training reduced curves more over 12 weeks; it did not measure position sense. A 2026 trial found small gains in trunk position sense, within measurement error, and no change in the curve over 16 weeks. The tests that would settle it have not been run yet.
The genetic evidence does not make scoliosis destiny. It points to a mechanism that may be partly neurological, not only structural. That is something that can be worked with, and measured.
I know the question firsthand. I had a scan and a diagnosis at 13, and at 18 a surgeon told me it was too risky to operate on and to just live with it. For the next 15 years I lived with it, as I had been told to.
When I started treating my scoliosis as a sensing problem, a nervous system pattern rather than only a bone defect, I gained roughly two inches of height, measured on a doorframe. No imaging was taken, so nobody knows what changed in my spine, or whether it did. My case sent me to the question; it is not evidence for the answer.
What conventional treatment misses
Standard scoliosis treatment works on the structure. Bracing applies outside force to slow curve progression during growth, and surgical fusion locks vertebrae in place with hardware.
Bracing works during the growth window. The research supports it. It is not designed to retrain the sensing system. Whether how a teenager perceives their own trunk when the brace comes off predicts later loss of correction is one of the tests the Neural Generation Hypothesis proposes.
Surgical fusion stabilizes the frame. It is not designed to change how the nervous system organizes the body above and below it.
Schroth and other scoliosis-specific exercise programs are the best-studied conservative option, and they already train breathing, position and awareness. If the sense of vertical is tilted, though, a patient may be trying to breathe into the concave side while their nervous system reads it as already level. That is a proposal worth testing, and it is where a sensory layer could add to what these methods do.
What the diagnosis misses
Traditional diagnosis tracks the shape, not the source. A Cobb angle tells you how curved the spine is. It tells you nothing about why it curved.
Two patients with the same Cobb angle may have different drivers. One may have an inner-ear asymmetry. Another may have a position-sense deficit. A third may have a bracing pattern that settled in during a growth spurt.
The curve can look the same on the X-ray even if the drivers differ. The hypothesis suggests adding a second question to assessment: not only “how big is the curve?” but “what is holding it?”, and that is testable.
The jaw connection
One more piece. In a 2023 review, people with idiopathic scoliosis (mean age about 14) were about twice as likely to have a crossbite or a Class II bite, and people with malocclusion had higher odds of scoliosis. An earlier review rated most such studies as weak and found no evidence that orthodontic treatment helps the spine, and uneven bites are also more common in scoliosis present from birth.
So the direction is open. The jaw is densely wired and feeds a lot of position information to the brainstem, close to the balance centers. An uneven bite may add to the sensory picture in which a curve is maintained. No study shows that it causes scoliosis in people.
What the genes do, and don’t, tell us
The genome is not a marketing department. It records what the body is built from, and so far it says scoliosis risk comes from both the structure and the sensing.
What’s shown, and what isn’t
- Shown: scoliosis runs in families (twin data put its heritability at about 38%). Risk genes sit on the structural side (PAX1, GPR126) and near a sensing gene (LBX1). Their effects are small.
- Shown in animals: removing the balance organ on one side in frogs, or the pineal gland (a hormonal change) in chickens, can produce scoliosis. Pineal removal did not do it in monkeys.
- Seen in people, measured only after the curve appeared: differences in inner-ear shape, the sense of vertical, and how the brain weights balance signals, in some patients. These could be a result of the curve rather than its cause.
- Not yet known: whether the human risk at LBX1 or POC5 works through sensing, and whether training the sensing system changes a curve.
That is the honest state of it. A clear lead, and the studies that would test it.
The curve is real. The suffering is real. The effects on breathing, movement and quality of life are real. But part of the source may sit in how the nervous system perceives gravity, processes position, and organizes the body around its best guess at vertical. That is what the next studies need to test.
The genes point beyond the spine. And that is good news.
If this changes how you think about your own scoliosis, or about posture in general, you are not alone. We work with people every day who spent years treating the shape before they looked at what was holding it.
Our scoliosis programming starts with the sensory-motor layer: balance, position sense, and how the body weights its signals. It works alongside standard care, not instead of it.
Sources
- Grauers A, Rahman I, Gerdhem P. Heritability of scoliosis. Eur Spine J. 2012;21(6):1069-1074.
- Takahashi Y, Kou I, Takahashi A, et al. A genome-wide association study identifies common variants near LBX1 associated with adolescent idiopathic scoliosis. Nat Genet. 2011;43(12):1237-1240.
- Cao Y, Min J, Zhang Q, Li H, Li H. Associations of LBX1 gene and adolescent idiopathic scoliosis susceptibility: a meta-analysis based on 34,626 subjects. BMC Musculoskelet Disord. 2016;17:309.
- Gross MK, Dottori M, Goulding M. Lbx1 specifies somatosensory association interneurons in the dorsal spinal cord. Neuron. 2002;34(4):535-549.
- Satoh D, Pudenz C, Arber S. Context-dependent gait choice elicited by EphA4 mutation in Lbx1 spinal interneurons. Neuron. 2016;89(5):1046-1058.
- Patten SA, Margaritte-Jeannin P, Bernard JC, et al. Functional variants of POC5 identified in patients with idiopathic scoliosis. J Clin Invest. 2015;125(3):1124-1128.
- Hassan A, Parent S, Mathieu H, et al. Adolescent idiopathic scoliosis associated POC5 mutation impairs cell cycle, cilia length and centrosome protein interactions. PLoS One. 2019;14(3):e0213269.
- Grimes DT, Boswell CW, Morante NF, Henkelman RM, Burdine RD, Ciruna B. Zebrafish models of idiopathic scoliosis link cerebrospinal fluid flow defects to spine curvature. Science. 2016;352(6291):1341-1344.
- Machida M, Dubousset J, Imamura Y, Iwaya T, Yamada T, Kimura J. Role of melatonin deficiency in the development of scoliosis in pinealectomised chickens. J Bone Joint Surg Br. 1995;77(1):134-138.
- O’Kelly C, Wang X, Raso J, et al. The production of scoliosis after pinealectomy in young chickens, rats, and hamsters. Spine. 1999;24(1):35-43.
- Cheung KM, Wang T, Poon AM, et al. The effect of pinealectomy on scoliosis development in young nonhuman primates. Spine. 2005;30(18):2009-2013.
- Lambert FM, Malinvaud D, Glaunès J, Bergot C, Straka H, Vidal PP. Vestibular asymmetry as the cause of idiopathic scoliosis: a possible answer from Xenopus. J Neurosci. 2009;29(40):12477-12483.
- Lambert FM, Malinvaud D, Gratacap M, Straka H, Vidal PP. Restricted neural plasticity in vestibulospinal pathways after unilateral labyrinthectomy as the origin for scoliotic deformations. J Neurosci. 2013;33(16):6845-6856.
- Hitier M, Hamon M, Denise P, et al. Lateral semicircular canal asymmetry in idiopathic scoliosis: an early link between biomechanical, hormonal and neurosensory theories? PLoS One. 2015;10(7):e0131120.
- Carry PM, Duke VR, Brazell CJ, et al. Lateral semi-circular canal asymmetry in females with idiopathic scoliosis. PLoS One. 2020;15(4):e0232417.
- Cortés-Pérez I, Salamanca-Montilla L, Gámiz-Bermúdez F, Obrero-Gaitán E, et al. Vestibular morphological alterations in adolescent idiopathic scoliosis: a systematic review of observational studies. Children (Basel). 2022;10(1):35.
- Pialasse JP, Descarreaux M, Mercier P, Simoneau M. Sensory reweighting is altered in adolescent patients with scoliosis: evidence from a neuromechanical model. Gait Posture. 2015;42(4):558-563.
- Cyr JP, Crepin R, Mercier P, Blouin JS, et al. Enhanced vestibular-evoked balance responses in adolescents with idiopathic scoliosis. J Neurophysiol. 2025;134(3):856-865.
- Cakrt O, Slabý K, Viktorinová L, Kolář P, et al. Subjective visual vertical in patients with idiopatic scoliosis. J Vestib Res. 2011;21(3):161-165.
- Kučerová K, Balatková Z, Slabý K, Rampová M, et al. Vestibular functions of adolescents with idiopathic scoliosis: a comprehensive assessment and comparative study. J Manipulative Physiol Ther. 2025;48(1-5):304-310.
- Wang D, Tsang R, Li Q, Chen F, Vasanthi RK, Purushothaman VK. Sensory integration dysfunction in adolescent idiopathic scoliosis: insights from a clinical balance assessment using the m-CTSIB. J Clin Neurosci. 2025;136:111267.
- Gámiz-Bermúdez F, Ibáñez-Vera AJ, Obrero-Gaitán E, Cortés-Pérez I, Zagalaz-Anula N, Lomas-Vega R. Relationship between stomatognathic alterations and idiopathic scoliosis: a systematic review with meta-analysis of observational studies. EFORT Open Rev. 2023;8(10):771-780.
- Langella F, Fusini F, Rossi G, Villafañe JH, Migliaccio N, Donzelli S, Berjano P. Spinal deformity and malocclusion association is not supported by high-quality studies: results from a systematic review of the literature. Eur Spine J. 2019;28(7):1638-1651.
- Zhang H, Ma J, Zhang Z, Feng Y, Cai C, Wang C. Occlusal deviations in adolescents with idiopathic and congenital scoliosis. Korean J Orthod. 2022;52(3):165-171.
- Liang W, Zhou Z, Lv Y, Zhou W, Wang H. Effect of sensory integration training combined with Schroth therapy on adolescents with mild idiopathic scoliosis: a 12-week randomized controlled trial. Eur J Med Res. 2025;31(1):59.
- Yaylacı SN, Demirkıran HG, Yağcı G. Integrating basic body awareness therapy with scoliosis-specific exercise for trunk proprioception in adolescent idiopathic scoliosis: a randomized controlled trial. Eur Spine J. 2026. doi:10.1007/s00586-026-10340-5.
- Miller SA. The Neural Generation Hypothesis of Adolescent Idiopathic Scoliosis. Version 3.0. Preprint, Zenodo, 2026 (not peer reviewed). doi:10.5281/zenodo.23685652.
Related research
- Not Everyone Has The Gene. Here’s What’s Actually Generating the Curve.
- Why 80% of Scoliosis Has No Explanation (Until Now)
- Schroth vs Functional Patterns vs Somatic Exercises: Which Works for Scoliosis?
- Functional Patterns for Scoliosis: What It Gets Right and What It Misses
Comments
2 responses to “Is Scoliosis Genetic? The Genes Point Beyond the Spine”
[…] Is Scoliosis Genetic? The Genes Point Beyond the Spine […]
[…] Is Scoliosis Genetic? The Genes Point Beyond the Spine […]