Amid advances in DNA testing, exams and radiographs still best for AIS diagnosis

The early diagnosis of adolescent idiopathic scoliosis in pediatric patients through radiographic studies or physical examinations can jumpstart treatment, but there is growing evidence, according to sources who spoke with Spine Surgery Today, that in some cases traditional conservative treatment may not help improve a scoliotic curve. Therefore, researchers and physicians continue to work on developing a genetic test or procedure to help determine which children will experience worsening of their curves. The hope is that curve progression may be halted and multiple invasive surgeries can be avoided.

Image: Marty Perlman

Observation and bracing are among the conservative options spine deformity specialists can consider. However, other factors are known to affect whether a scoliosis curve will worsen or remain the same.

According to Carol A. Wise, PhD, director of molecular genetics at Texas Scottish Rite Hospital for Children in Dallas, the “holy grail” for surgeons and researchers is the development of a test that will accurately predict which patients will have progressive or non-progressive adolescent idiopathic scoliosis (AIS). There is also a need to understand what causes these forms of idiopathic scoliosis.

For example, Wise and colleagues are investigating the PAX1 gene. She said it is important in early development of the spine.

“There is also a gene called GPR126. We have shown in our labs, and other labs have shown, there are clearly variations in this gene that can increase susceptibility to AIS. One of the most exciting things about this gene is recently a research group made a conditional knockout, which means they removed the gene specifically in the developing cartilage of mice, and the mouse developed what looks like AIS. The reason that is thrilling and exciting is because we have never had a true model of AIS. This model will become an important research tool,” Wise told Spine Surgery Today.

Essential to a treatment plan

Amid new findings surrounding the genetic aspects of AIS, the accurate diagnosis of idiopathic or non-idiopathic scoliosis should be the first step for any physician when determining a treatment plan for a patient with scoliosis, according to David L. Skaggs, MD, MMM, Chief of the Department of Orthopaedic Surgery, Children’s Hospital Los Angeles, in Los Angeles.

“I think a spine surgeon caring for a child or teenager has to first establish whether it is idiopathic. If it is not idiopathic, there could be another underlying, treatable cause, such as a tethered cord, a non-malignant tumor, like an osteoid osteoma, or cancer,” Skaggs told Spine Surgery Today. “Although it sounds old-school, history or a physical examination can determine if a curve is idiopathic. A good way is to teach residents or fellows going into the field to think of themselves as a detective. You want to look at as many clues as you can to determine if a curve looks idiopathic.”

Idiopathic curve detection

An idiopathic curve can be seen in the frontal plane visible in a front-to-back radiograph. The spine also shows clear rotation. When a patient bends forward, a rib asymmetry can be seen in those patients who have idiopathic curves, Skaggs said.

If a patient does not exhibit this rib asymmetry or rotation, or complains of abnormal pain wakes up at night due to pain, or has pain running down his or her arms and legs, it suggests that their curve is not idiopathic, Skaggs said.

Once you have a diagnosis of AIS, Skaggs said the next step is to determine if a patient’s will progress or not.

“There are two factors on how to decide if it progresses,” Skaggs said, which are the size of the curve and the state of the patient’s maturation, or how much growth he or she has left.

“Size of the curve is straightforward. Right now, we use Cobb angle. I predict in the future we will be using other things, such as rotation and perhaps total 3-D deformity, looking at maybe the amount of hypokyphosis in the thoracic curve,” he said.

Measure, check radiographs

Spine surgeons can use several measurements to determine the state of maturation and growth remaining in a patient, according to Paul D. Sponseller, MD, MBA, director of pediatric orthopedics at Johns Hopkins Children’s Center in Baltimore.

“We look at radiographic predictors of growth remaining, such as the digital hand atlas. We use other radiographic views, such as elbow X-rays and foot X-rays. We also look at cartilage ratings of the hip. All of those can predict if a kid has a lot of growth remaining or is near the end of their growth,” Sponseller told Spine Surgery Today.

Christopher I. Shaffrey, MD, FAANS, of the Department of Neurological Surgery at the University of Virginia Health System in Charlottesville, Va., said frequency of the examination depends on the age of the child or adolescent, the level of skeletal maturity and the severity of the curve at diagnosis.

“For very mild curves, observation may be all that is needed. For scoliotic curves greater than 20° [in patients who] are skeletally immature and have a high rate of progression, bracing is indicated. For curves greater than 50° at diagnosis or those [patients who] progress to 50° with bracing, surgical management of the scoliosis with instrumentation and fusion is indicated,” Shaffrey told Spine Surgery Today.

One test, the ScoliScore AIS Prognostic Test, is a DNA-based prognostic test designed to help predict the risk of curve progression in patients with AIS. However, an independent study conducted by researchers at Columbia University and published in Journal of Bone & Joint Surgery in 2015 showed scores on this test did not differ between patients with and without the curve progression.

The final analysis of the 126 patients studied revealed no significant differences in rates of progression in patients with the highest risk ScoliScores compared with patients with the lowest risk scores.

Feedback on testing, bracing

Skaggs used the ScoliScore test when it first became commercially available.

“After using it for about a year, there was never one case in which it predicted if a curve would get worse or not that beat my clinical judgement. I did not think it added anything. The cost was about $2,000 a patient, so I stopped using it,” Skaggs said.

A definitive DNA or genomic-based test could reveal which patients are genetically predisposed to worsening of their curve and which patients would benefit from bracing treatment, Skaggs said.

He noted that bracing is an effective treatment for AIS patients who are at high risk for curve progression, based on the results of a study in the New England Journal of Medicine in 2013. The multicenter study found bracing significantly decreased the progression of high-risk curves to the threshold for surgery in AIS patients.

“If it is a genetic disease, and I believe it has been proven to be, of course there could be a future genetic test that may be beneficial and may say who would benefit from a brace. If we know a brace changes the chances of one in three kids needing surgery, perhaps the first step will be determining which kids will benefit from a brace,” Skaggs said.

Genetic research progresses

The human genome has 3 billion base pairs, according to Wise. She said she and her colleagues are looking at millions of base pairs at a time in patients with and patients without AIS. They analyze the polymorphic markers and identify gene variants in AIS patients compared with control patients to find any differences in their genomes.

“[Statistically], we look at these variants occurring far more frequently in our cases compared with our controls, or vice versa. Statistically, we see this is against the odds, so this must be marking a region of the genome that is contributing to the disease... [In] doing this, we and other groups have located parts of the genome that clearly increase the risk of AIS,” Wise said.

AIS is an adolescent-onset disease. Thus far, Wise and colleagues know the PAX1 gene plays an important role in early developmental stages of the spine so they are designing studies to test if PAX1 has a role later on when a patient is developing AIS or if there is another factor that predisposes patients to AIS and pushes them over the edge to develop this deformity.

Wise said she and her colleagues are likewise studying the gene LBX1. It has an important role in the early development of muscle and may therefore be important to the development of AIS and curve progression in a patient later on, she said.

Studying patients with the most severe AIS to identify what makes them different in terms of their genetic makeup may lead researchers to develop a possible genetic test that will show which patients will do well with bracing and which patients are predisposed to worsening of their curves, Wise said.

Surgery during growth spurt

Peter O. Newton, MD, told Spine Surgery Today there are several surgical techniques that can be used when a patient is still growing that could provide similar beneficial outcomes compared with spinal fusion, but without the potential downside of fusion.

“There are growth modulation techniques that are in the surgical realm that do not involve spinal fusion that are beginning to be utilized. The experience [with these] still remains relatively limited, with follow-up all less than 5 years. But, clearly, there are growth-modulating techniques that have been utilized to change the way the spine grows that have the potential to correct scoliosis without having to perform spinal fusion,” Newton said. “These methods typically involve an anterior intersegmental screw fixation with some flexible element connected between the screws on the convex side of the curve, which limits convex anterior growth and allows the patient to grow out of their scoliosis.”

This is off-label use of the surgical implants involved, but preliminary assessment has shown they have early promise in treating a patient’s curve, he said. – by Robert Linnehan

• References:
• Roye BD, et al. J Bone Joint Surg Am. 2015;doi:10.2106/JBJS.O.00217.
• Weinstein SL, et al. N Engl J Med. 2013;doi:10.1056/NEJMoa1307337.

• For more information:
• Peter O. Newton, MD, can be reached at Pediatric Orthopedic and Scoliosis Center, 3030 Children’s Way, Suite 410, San Diego, CA 92123; email: cmcginley@rchsd.org.
• Christopher I. Shaffrey, MD, FAANS, can be reached at P.O. Box 800386, Charlottesville, VA 22908; email: cis8z@hscmail.mcc.virginia.edu.
• David L. Skaggs, MD, MMM, can be reached at Children’s Hospital of Los Angeles, Department of Orthopedic Surgery, 4650 Sunset Blvd., MS69, Los Angeles, CA 90027; email: dskaggs@chla.usc.edu.
• Paul D. Sponseller, MD, MBA, can be reached at Johns Hopkins Children’s Center, 601 N. Caroline St., 5th Floor, Baltimore, MD 21287; email: psponse@jhmi.edu.
• Carol A. Wise, PhD, can be reached at Texas Scottish Rite Hospital for Children, 2222 Welborn St., Dallas, TX 75219; email: carol.wise@tsrh.org.

Disclosures: Newton reports he is a paid consultant to Cubist, receives research support from DePuy Synthes Spine via Setting Scoliosis Straight Foundation, receives IP royalties from and is a paid consultant or presenter for DePuy Synthes Spine and has stock or stock options with ElectroCore, he receives research support from EOS Imaging, is a board/committee member of the International Pediatric Orthopedic Think Tank, is a paid consultant to K2M and paid speaker/presenter for K2M and receives research/institutional support from K2M via the Setting Scoliosis Straight Foundation, he receives research support from Orthopaediatrics, is a committee member of the Scoliosis Research Society (SRS) and the Setting Scoliosis Straight Foundation and receives publishing royalties, financial or materials support from Thieme Publishing. Shaffrey reports he is a consultant to Zimmer Biomet, Medtronic, NuVasive, K2M and Stryker, he receives royalties from patents with Zimmer Biomet, Medtronic and NuVasive and receives grants from the NIH, U.S. Department of Defense, North American Clinical Trials Network, AO, Neurosurgery Research and Education Foundation, DePuy Synthes, and International Spine Study Group. Skaggs reports his institution receives grants from POSNA and from Ellipse paid to Growing Spine Foundation (GSF); receives consulting fees from Zimmer Biomet, Medtronic, Zipline Medical Inc., Orthobullets and Grand Rounds, he has stock options with Zipline Medical Inc., has board membership on the Growing Spine Study Group, SRS and GSF and is paid for speaking/educational lectures by Zimmer Biomet, Medtronic and Johnson & Johnson; He has patents with Medtronic and Biomet and receives royalties from Wolters Kluwer Health – Lippincott Williams & Wilkins, and Biomet Spine. Sponseller and Wise report no relevant financial disclosures.

Is the ScoliScore the most effective test to predict the onset of adolescent idiopathic scoliosis?

 

Small role for single nucleotide polymorphisms

There is sufficient published evidence the ScoliScore has limited clinical utility as a predictor of spinal curve progression. The ScoliScore is inherently flawed because it evaluates one class of genetic markers to make a prediction of spinal curve progression. In addition to the common single nucleotide polymorphisms that make up the ScoliScore, a vast amount of genetic difference between individuals consists of rare and private single nucleotide variants, variably sized deletions and duplications of the genome and epigenetic marks.

Common single nucleotide polymorphisms (SNP) c+++++an be strongly linked to the risk for disease, as has been shown for Alzheimer’s disease and the AP0E4 allele. A single APOE4 allele increases risk 3.2 fold and two alleles increases risk of Alzheimer’s by nearly 15-fold, so SNPs are clearly important for some diseases. SNPs are present in the population at high frequency (greater than 5%) and millions can be detected inexpensively on a single microarray, therefore this technology drove the era of the genome-wide association study (GWAS) on which the ScoliScore is based. However, for many diseases, including adolescent idiopathic scoliosis (AIS), SNPs play only a small role.

With advances in sequencing technology, the $1000 Genome is on the horizon, and studies are realizing the genetic architecture of many diseases is strongly driven by rare nucleotide differences, not by the SNPs that could be studied in GWAS studies. Large-scale sequencing studies to identify rare variants associated with AIS are in their infancy and need to be completed before the contribution of all of these genetic differences to AIS is fully understood.

While common genetic variants tested by the ScoliScore are of limited clinical utility, the truth is we do not yet know whether even a comprehensive set of common and rare genetic markers will be effective in the prediction of spinal curve progression. Genetics research currently in progress will provide answers, perhaps revealing a stronger component of environmental factors than ever imagined.

Matthew B. Dobbs, MD, is professor and director of strategic planning, Department of Orthopaedic Surgery, Washington University School of Medicine, in St. Louis.

Disclosure: Dobbs reports no relevant financial disclosures.

 

Not widely applicable for high-risk patients

ScoliScore is a 53-gene panel developed with the intent of estimating the risk of curve progression to surgery in patients with AIS, assigning a numeric risk score between 1 and 200. In the index study, low scores of 1 to 50 were associated with a low risk of curve progression, while a score of 181 to 200 was associated with a 50% to 99% chance of progression to a Cobb angle greater than 40°. The aspiration was that this test would provide a reliable way to both identify those children at low risk for curve progression (thus reducing costs, as well as radiation exposure) and those children at high risk for curve progression who would potentially benefit from early intervention. Indications for ScoliScore testing include a diagnosis of AIS, a Cobb angle between 10° and 25°, skeletal immaturity and Caucasian race.

In their initial study in 2012, Roye and colleagues suggested the ScoliScore was useful to stratify risk in skeletally immature children with curves greater than 20°, but the group’s larger subsequent study in 2015 revealed no difference between patients with and those without curve progression.

Several more studies attempting to replicate the original results in differing ethnic populations (Ogura and colleagues in 2013, Tang and colleagues in 2015 and Xu and colleagues in 2016) have failed to demonstrate an association of the ScoliScore gene panel with AIS curve progression in their patient populations.

Since the ScoliScore has not proven to be widely applicable for identification of children at highest risk of significant curve progression, the effort to refine genetic testing continues. Most surgeons continue to rely on assessment of clinical risk factors, some of which are becoming better defined. For example, the recent data from Karol and colleagues suggest a 75% rate of progression to surgery for R0 children who present with curves greater than 40° despite brace therapy. At present, no test reliably identifies the child at high risk for progression prior to onset of clinical deformity.

Laurel C. Blakemore, MD, is chief and associate professor of pediatric orthopaedics in the Department of Orthopaedics and Rehabilitation at University of Florida College of Medicine, in Gainesville, Fla.

Disclosure: Blakemore reports she has a consulting agreement with and receives research support from K2M Medical and is an associate editor of Spine Deformity.

• References:
• Karol LA, et al. J Bone Joint Surg Am. 2016;doi:10.2106/JBJS.O.00359.
• Ogura Y, et al. Spine. 2013;doi:10.1097/BRS.0b013e3182947d21.
• Roye BD et al. J Bone Joint Surg Am. 2015;doi:10.2106/JBJS.O.00217.
• Roye BD, et al. Spine. 2012;doi:10.1097/BRS.0b013e31825eb605.
• Tang QL, et al. Spine. 2015;doi:10.1097/BRS.0000000000000807.
• Xu L, et al. Spine. 2016;doi:10.1097/BRS.0000000000001203.