How should dose homogeneity across bone be handled in pediatric patients?

The radiation-associated bone damage can be affected by many factors, including total radiation dose, fractionation schedule, treatment volume, age of the child, symmetry of the delivered dose over vertebrae, developmental status of the irradiated growth plates, treatments such as chemotherapy or surgery, inherent factors such as endocrine abnormalities, and bone growth aberrations because of tumor location. The spectrum of skeletal effects of radiation therapy includes fracture, osteoradionecrosis, and growth retardation. However, in this discussion, we are focusing on how to prevent radiation-induced spinal asymmetrical growth only.

To minimize the likelihood of asymmetrical growth, when is it of sufficient concern to necessitate altering planning objects?

When a part of vertebral body receives ≥ 8 Gy, you need to treat the entire vertebral body.

Doses as low as 8 Gy in young children and 14 Gy in older children have been reported sufficient to cause growth retardation of the epiphysis.¹ The concept of partial vertebral body irradiation, even at low doses, remains controversial. Current COG ANBL1531 protocol recommends “Care must be taken to avoid a steep dose gradient that transverses the vertebral body. When any part of a vertebral body must be included in the PTV, the entire vertebrae (corpus, pedicles, transverse and spinous process) should receive at least 18 Gy. The target volume should not be modified to include the vertebral body; rather, the vertebral body should be contoured as an additional pseudo-target structure with a specified minimum dose of 18 Gy if any part of it is included in the PTV.”

When planning paediatric lymphoma case, how would you manage dose inhomogeneity in the spine?

Use field-in-field (poor man’s IMRT) or IMRT technique to keep dose gradients in the vertebral body as low as possible or < 5 Gy.

Studies of partial spinal radiotherapy report more frequent and more severe scoliosis after a long follow-up (at least 5–10 years) and high radiation doses (doses of >18 Gy or vertebral dose gradients of >10 Gy).^2-5 Dorr and colleagues² found a significant inverse relationship between age at radiotherapy treatment and the development of scoliosis. For vertebral body dose gradients of less than 35 Gy, especially when the dose gradients were between 5 Gy and 20 Gy, a steep increase in incidence of scoliosis was observed in children aged 6 years or younger at treatment, but not in older age groups. For children younger than 6 years of age, dose gradients within the vertebral body of 10 Gy, 20 Gy, and 30 Gy resulted in a 50%, 85%, and 100% incidence of substantial kyphoscoliosis, respectively. For gradients of more than 35 Gy, no further effect of dose was found; the effect of radiation on increasing the risk of scoliosis was already maximized at a gradient of 35 Gy.

Do you look for any specific isodose line that should cover entire spine?

Hodgkin lymphoma’s prescription dose is usually 21 Gy per COG trials, so I will cover the vertebral body with a minimum of 18 Gy isodose line.

What ages are relevant for this consideration? i.e. Are pediatric Hodgkin patients at risk?

Any kids who have not completed spine growth are relevant for this consideration. So, Hodgkin patients who are skeletally immature are at risk.

Are there studies which can help determine extent of epiphyseal closure and bone age?

The most widely used skeletal maturity assessment techniques used today are the Greulich-Pyle and Tanner-Whitehouse 2 approaches, both involving left hand and wrist radiographs.^6-9 Because the axial skeleton matures a few years later than the remaining skeleton, consideration of a more focused assessment for spinal maturity in select patients using the Risser sign may be warranted,¹⁰ which should be assessed by an expert pediatric radiologist. The Risser sign is an indirect measure of skeletal maturity, whereby the degree of ossification of the iliac apophysis by x-ray evaluation is used to judge overall skeletal development. Mineralization of the iliac apophyses begins at the anterolateral crest and progresses medially towards the spine.

What is the dose threshold for consideration of mitigating non-uniform doses across bone/vertebrae?

Up-to-date, there is no study showing “bone growth ablation” dose and beyond this dose, dose inhomogeneity should not be a concern because there is no dose response after this dose level. Acknowledging no prospective randomized data, full vertebral body coverage for skeletally immature patients was recommended for proton CSI.¹¹

Is there a trade-off between increased integral dose to minimize growth asymmetry and increased risk for second cancers?

Since growth asymmetry is a more common issue than secondary cancer and since we don’t know exactly the range of doses that will cause secondary cancer, it is worthwhile to minimize growth asymmetry.

1. Heaston DK, Libshitz HI, Chan RC. Skeletal effects of megavoltage irradiation in survivors of Wilms' tumor. AJR Am J Roentgenol. 1979;133:389-395.
2. Dorr W, Kallfels S, Herrmann T. Late bone and soft tissue sequelae of childhood radiotherapy. Relevance of treatment age and radiation dose in 146 children treated between 1970 and 1997. Strahlenther Onkol 2013; 189: 529–34.
3. Paulino AC, Fowler BZ. Risk factors for scoliosis in children with neuroblastoma. Int J Radiat Oncol Biol Phys 2005; 61: 865–69.
4. Makipernaa A, Heikkila JT, Merikanto J, et al. Spinal deformity induced by radiotherapy for solid tumors in childhood: a long-term follow up study. Eur J Pediatr 1993; 152: 197–200.
5. Sasso G, Greco N, Murino P, et al. Late toxicity in Wilms tumor patients treated with radiotherapy at 15 years of median follow-up. J Pediatr Hematol Oncol 2010; 32: e264–67.
6. Greulich WW, Pyle SI. Radiographic atlas of skeletal development of the hand and wrist. Am J Med Sci 1959;238:393.
7. Goldstein H, Cameron N, Healy JM, et al. Assessment of Skeletal Maturity and Predication of Adult Height (TW3 Method). London: WB Saunders; 2001.
8. Satoh M. Bone age: Assessment methods and clinical applications. Clin Pediatr Endocrinol 2015;24:143-152.
9. Bull RK, Edwards PD, Kemp PM, et al. Bone age assessment: A large scale comparison of the Greulich and Pyle, and Tanner and Whitehouse (TW2) methods. Arch Dis Child 1999;81:172-173.
10. Waldt S, Woertler K, Telger TC. Measurements and classifications in musculoskeletal radiology. Stuttgart, New York: Thieme; 2014.
11. McMullen K, Buchsbaum J, Douglas J, et al. Growth abnormalities of the spine after radiation therapy: Respecting the past while moving forward in proton craniospinal irradiation. Practical Radiation Oncology (2013) 3, 337–343.

To minimize the likelihood of asymmetrical growth, when is it of sufficient concern to necessitate altering planning objectives?
What ages are relevant for this consideration?

The SIOPE Radiotherapy working group recently published consensus recommendations for management of vertebral radiotherapy dose in pediatric patients (Hoeben et al. Lancet Oncology 2019)¹

For paravertebral tumors in patients not receiving CSI, they recommend dose gradients as follows (from Figure 2 of the manuscript):

• Age 0-2 years: < 3 Gy
• Age 2-6 years: < 3 Gy
• Age 6 to end of pubertal acceleration phase: < 5 Gy
• After end of pubertal acceleration phase: All gradients acceptable

They recommend
- avoiding a dose of > 20 Gy to 7 or more thoracic vertebrae
- establishing the end of the pubertal acceleration phase with the Risser and Sauvegrain methods

This manuscript includes a nice figure showing the primary and secondary ossification centers of different areas of the bone, which can help guide “elective bone coverage” especially when adjacent to another organ at risk (e.g. the kidney or heart), which can be sensitive to low to moderate doses.

In practice, using protons, I am more tolerant of a dose gradient in the AP/PA direction as opposed to the right-left direction. Even steep gradients in the AP/PA direction (e.g. from 0 Gy to full CSI dose of 23.4 or 36 Gy across the vertebral body) using vertebral-body-sparing proton-CSI in young patients has been shown in a small series at Loma Linda of 6 patients ages 3-5 to not result in clinical kyphosis or lordosis².

Furthermore, it is important to be cognizant of the trade off between bone deformity (likely a non-fatal toxicity) vs. other potentially late toxicity. For example, in treating curative pediatric patients, especially with Hodgkin Lymphoma, I try to keep the 5 Gy line off of all cardiac structures if possible, a bit more conservative than the ILROG adult mediastinal lymphoma guidelines (Dabaja 2018), which recommend mean < 5 Gy to the heart and heart substructures.³

In a recent survey of practice patterns of physicians delivering proton-CSI (Medek et al. 2019), when physicians were shown 3 treatment plans of a 7 year old (skeletally immature) child receiving CSI, the most commonly chosen plan included (only) 80% isodose line coverage of the vertebral body with 20% isodose line extending to the posterior of the heart. Whereas an alternative plan with 95% isodose line covering the entire vertebral body with 20% isodose line extending mid-way into the heart was chosen by a smaller proportion.⁴

As such, please remember the guidelines are to be used exactly as such, and clinical discretion is paramount when electively irradiating a non-target area.

What is the dose threshold for consideration of mitigating non-uniform doses across bone/vertebrae?

The SIOPE guidelines (in Figure 2 of manuscript) quote as follows:
“For a target dose > 40 Gy, higher gradients might be allowed, with a minimum vertebral dose of 36 Gy.”

When planning paediatric lymphoma case, how would you manage dose inhomogeneity in the spine?
Do you look for any specific isodose line that should cover entire spine?

Most commonly I am treating ISRT in an adolescent patient with a bulky mediastinal mass + any slow responding lesions (if any), as per the Children's Oncology Group.

Given that mediastinal irradiation tends to cover the spine from Left-to-Right, and patients tend to be older, gradients are not as much an issue as in other pediatric histologies.

However, if treating a younger patient with Hodgkin Lymphoma to 21 Gy, I would aim for ~18 Gy coverage as per ANBL1531 neuroblastoma protocol, which uses 21.6 Gy prescription for all patients or potentially up to a 5 Gy gradient as per the SIOPE recommendations. If going to a higher dose (e.g. 30 Gy), I would likely minimize gradient to 5 Gy.

Again, however, I would prioritize minimizing cardiac dose and meeting other organ at risk dose constraints over elective bone irradiation when faced with the decision.

REFERENCES

1. Hoeben et al.  Management of vertebral radiotherapy dose in paediatric patients with cancer: consensus recommendations from the SIOPE radiotherapy working group. Lancet Oncology 2019.
2. MacEwan et al. Effects of vertebral-body-sparing proton craniospinal irradiation on the spine of young pediatric patients with medulloblastoma. Adv Radiat Oncol 2017.
   
   
3. Dabaja et al. Proton therapy for adults with mediastinal lymphomas: the International Lymphoma Radiation Oncology Group guidelines. Blood 2018.
   
   
4. Medek et al. Practice Patterns Among Radiation Oncologists Treating Pediatric Patients With Proton Craniospinal Irradiation. PRO 2019.