Palliative radiotherapy in pediatrics


Background

An estimated 11,060 new cases of pediatric cancer will be diagnosed in the United States in 2019. In children ages 0 to 14, the most common diagnoses are leukemia (28%), central nervous tumors (21%), and neuroblastoma (7%) ( Fig. 17.1 ). In adolescents ages 15 to 19 the most common malignancies are Hodgkin lymphoma (15%), thyroid carcinoma (11%), and central nervous system (CNS) tumors (10%). The survival rates for pediatric malignancies have significantly improved from approximately 63% in the mid-1970s to 83% today. However, cancer remains the second leading cause of mortality in children after accidental death, and an estimated 1190 children will die of their disease in 2019. Long term survival varies widely by diagnosis, ranging from 98% overall in patients with Hodgkin lymphoma to less than 2% in patients with diffuse intrapontine glioma (DIPG). ,

Fig. 17.1, Childhood cancer incidence.

For the majority of children with cancer, the goal of treatment is cure. A significant proportion of high risk patients are treated on experimental protocol therapy, early phase clinical trials, and receive hematopoietic stem cell transplants. Approximately 30% to 50% will definitive receive radiation therapy (RT) during their disease course. , Aggressive, curative intent therapy may be delivered even in the setting of metastatic or recurrent disease and many patients receive cancer-directed therapy during the last month of life. ,

Throughout the course of treatment, children may experience symptoms which can result in pain, other types of suffering, and decreased quality of life. Children with leukemia and lymphoma, older patients, and patients undergoing stem-cell transplantation are at high risk for poorly controlled symptoms, but cancer-related symptoms can be significant in any child receiving end of life care for progressive malignancy. In a large single institution series examining pediatric cancer patients and families receiving end-of-life care, 89% expressed a great deal of suffering from at least one symptom, and 51% from 3 or more symptoms. Commonly reported symptoms included fatigue, pain, dyspnea, and poor appetite. Another study reported that 94% suffered from at least 3 symptoms while 76% experienced at least 5 symptoms at the end of life. Parental quality of life scales have correlated with the level of child anxiety and pain at the end of life, suggesting that integration of palliative care may both reduce patient suffering and improve long-term family wellbeing.

Assessing symptoms and distress may require age-appropriate communication, assessment tools, and baseline parental assessments. , While older patients may be able to easily describe symptoms, younger patients may demonstrate pain, anxiety, and treatment-related discomfort only through withdrawal and decreased activity. Pediatric specific symptom assessments may be useful for clinicians assessing baseline symptoms and treatment response when parents and/or children are able to complete them. ,

Children with high-risk cancer and their families experience significant physical, psychological, social, and spiritual challenges. Integration of pediatric palliative care (PPC) into the oncology care team has become more common over the past two decades and is associated with improved patient and family quality of life. The World Health Organization (WHO) defines PPC as therapy directed to address the needs of children with life-threatening illnesses through “the active total care of the child’s body, mind, and spirit, and support to the family. PPC begins when illness is diagnosed and continues regardless of whether or not a child receives treatment directed at the disease.” Goals of PPC particularly for children with cancer may include symptom management, integrative therapies, counseling, support services, and advanced care planning.

Palliative RT may be used to treat life or organ threatening emergencies and for control or prevention of symptoms in pediatric patients with cancer ( Table 17.1 ). Of all children receiving radiation, approximately 11% to 18% will receive radiation treatment with palliative intent. However, reporting is variable across institutions, highlighting the range of definitions of palliative intent especially in the pediatric population ( Table 17.2 ). While numerous large, prospective studies of dose and fractionation exist for palliation of brain metastases, bone metastases, and cord compression in adult patients, data in pediatric patients are limited and practice is often extrapolated from adult palliative literature.

TABLE 17.1
Common Indications for Palliative Radiation in a Multi-institutional Series
Data from Rao AD, Chen Q, Ermoian RP, et al. Practice patterns of palliative radiation therapy in pediatric oncology patients in an international pediatric research consortium. Pediatr Blood Cancer . 2017;64(11).
Pain 43%
Intracranial symptoms 23%
Respiratory compromise 14%
Cord compression 8%
Abdominal distention 6%
Bowel obstruction 3%
Postoperative spine 2%

TABLE 17.2
Definitions of Palliative Radiation
Rao et al. Treatment with the goal to improve symptoms or prevent impending symptoms.
Mak et al. Treatment of advanced, incurable cancer. Patients enrolled on protocols that called for irradiation of metastases present at diagnosis were excluded.
Varma et al. Treatment of patients with advanced cancer who were ineligible for or whose disease had persisted/progressed through standard of care first-line therapy and in whom the goal of RT was amelioration or prevention of a specific symptom.
Rahn et al. Treatment of patients who were symptomatic and believed to have incurable disease at the time of treatment.
Lazarev et al. Treatment of patients with metastatic or recurrent disease.
RT , Radiation therapy.

Patient age and comfort with treatment must also be considered to safely and accurately deliver palliative RT. For very young patients, anesthesia may be required. While the majority of children 3 years old or younger require anesthesia, its use may be minimized in older patients. Orientation to the RT department, reassurance, behavioral modification, and educational, child-oriented videos may reduce the need for anesthesia. Child life therapists are also a beneficial resource to introduce the child to the simulation and treatment rooms, familiarize them with set-up material required, and to coach them on exercises or play therapy that may make them feel more comfortable with the experience.

Effective communication is important to identify both the patient and family’s understanding of the disease, willingness to tolerate side effects, and anticipated treatment outcome prior to delivery of palliative RT ( Table 17.3 ). , Patients as young as 3 years of age may be aware of their disease process and prognosis, and their level of understanding and involvement in treatment decisions should be discussed with their families and care teams. , Parents of children with cancer have complex responses to treatment recommendations and may have difficulty comprehending curative versus symptom-directed care. In a questionnaire to parents of patients receiving palliative RT, 76% anticipated improvement in quality of life after palliative RT. An equal number (76%) also expected prolongation of life and 40% expected palliative RT to cure their child’s malignant disease. Clear discussion with the patient, family, primary oncologist, PPC, and hospice teams are critical in establishing prognosis and goals of RT prior to treatment.

TABLE 17.3
Exploring Goals of Care
Adapted from Waldman E, Wolfe J. Palliative care for children with cancer. Nat Rev Clin Oncol . 2013;10(2):100–107.
Who is your child (as a person)?
What is your understanding of your child’s illness? What does the illness mean to you and your family?
In light of your understanding, what is most important regarding your child’s care?
What are your hopes for your child? What are your fears and concerns regarding your child?
Where do you find support and strength?

Solid malignancies

Emergencies

Oncologic emergencies generally present earlier in the course of disease in children, frequently as the presenting symptom, and do not always carry the poor prognosis associated with adult oncologic emergencies. Given the possible late complications of any cancer therapy, multidisciplinary discussion regarding prognosis, goals of therapy, and functional outcome is important to determine the optimal treatment.

Spinal cord compression

Spinal cord compression (SCC) is a frequent complication in pediatric oncology patients. It is estimated that 3% to 25% of patients will develop SCC at some time during their disease course. SCC may be the presenting symptom, reported in 30% to 67% of cases, and may also occur in the terminal phases of disease. Soft tissue sarcomas account for approximately 43% to 65% of SCC cases in children. , Neuroblastoma may present with cord or nerve root compression in 7% to 15% of cases and is a common cause of SCC in patients under the age of 5 years. Other less frequently observed histologies include Hodgkin lymphoma, non-Hodgkin lymphoma, and renal tumors. , ,

Back pain and lower extremity weakness are the most common presenting symptoms. More advanced cases may present with sphincter dysfunction including urinary retention or constipation. Infants may present with more subtle symptoms including irritability, often leading to a delay in diagnosis. , , These young patients may also have symptoms less frequently seen in adult patients with SCC including visible flank masses and respiratory compromise.

Steroids are generally initiated urgently for symptomatic control, although may be delayed for patients without a histologic diagnosis. , Magnetic resonance imaging (MRI) is the most sensitive diagnostic imaging and is recommended if it can be obtained emergently. For those patients without a tissue diagnosis, laminectomy may be required if there are no other sites of disease amenable to biopsy or resection. ,

Treatment options for SCC in pediatric patients include decompressive surgery, systemic therapy, and RT. Decisions regarding optimal management are complex and depend on the patient’s stage, prior therapy, histology, degree of symptoms, and prognosis.

Cord compression at initial diagnosis.

For patients with potentially curable disease who present with cord compression at the time of diagnosis, future need for definitive RT and late toxicities of treatment should be considered ( Fig. 17.2 A). For chemosensitive histologies including Ewing sarcoma or rhabdomyosarcoma where radiation may be indicated in the future to therapeutic doses, it is important to avoid exceeding dose tolerances by delivering palliative treatment in the upfront management. In these cases, starting with chemotherapy often provides a therapeutic response and is recommended as first line therapy in newly diagnosed, responsive histologies.

Fig. 17.2, Management of spinal cord compression (SCC). (A) Management of SCC in the primary setting. (B) Management of SCC in the recurrent setting.

For patients with high grade neurologic deficits and chemosensitive histologies laminectomy may be considered ( Table 17.4 ). , However, extensive surgery may result in significant late effects which should be considered for patients presenting with cord compression who have received no other therapy. In surgical series, 31% to 62% of patients experienced postoperative scoliosis, 26% to 34% had kyphosis, and severe deformities requiring spinal fusions were reported in 28%. More recent series demonstrate spine abnormalities in 52.9% of patients treated by neurosurgery compared to 23.5% in patients treated with chemotherapy alone.

TABLE 17.4
Grading of Neurologic Dysfunction
Data from Gilbert RW, Kim JH, Posner JB. Epidural spinal cord compression from metastatic tumor: diagnosis and treatment. Ann Neurol . 1978;3(1):40–51.
Grade 1 Ambulatory patient with or without hypotonia or ataxia of the legs
Grade 2 Nonambulatory patient, but able to raise the legs against the gravity force while laying down
Grade 3 Unable to raise the legs against the gravity force or paraplegia

In contrast, laminectomy in the upfront setting is indicated for resistant histologies such as non-rhabdomyosarcoma, non-Ewing soft tissue sarcoma or bone sarcoma. Klein et al. demonstrated that in patients with non-rhabdomyosarcoma, non-Ewing soft tissue sarcoma, surgically treated patients had a higher average posttreatment functional score compared to those treated medically. In this series, of 22 patients with a pretreatment symptomatic cord compression, 86% improved and 17% had a complete response. Pollono et al. evaluated 70 patients with cord compression, 17 with soft tissue sarcoma. They recommended laminectomy for all patients with paraplegia except in cases of complete paraplegia for greater than 24 to 96 hours, recurrent disease, or poor prognosis. Laminectomy was also recommended in cases with chemoradiotherapy non-responsive tumors. Improvement was noted in 57% of all cases after laminectomy for paraplegia.

If RT is required for refractory cord compression from the primary tumor at the time of diagnosis, it is reasonable to consider treating to a definitive dose when delivery of emergent radiation would compromise the ability to give the curative dose in the future. This is of greatest concern for histologies such as Ewing or rhabdomyosarcoa, where the definitive dose is at the tolerance of the spinal cord. For other histologies such as neuroblastoma where cord tolerance is of less concern at the definitive dose, low-dose palliative radiation prior to definitive local therapy may be considered.

Cord compression in the relapsed/refractory setting.

For patients presenting with cord compression in the relapsed/refractory setting, management is similarly guided by histology with less concern for possible future definitive treatment ( Fig. 17.2 B). Patients who have chemosensitive histologies and have systemic therapy options may be treated upfront with chemotherapy. For heavily pretreated patients not expected to respond rapidly to systemic therapy or those without systemic therapy options, emergent RT may provide a rapid response and durable local control. In the setting of chemoradiation resistant histologies where surgery would not be significantly morbid, resection may provide the best functional outcome.

RT doses in the palliative setting vary widely, largely based on histology. Holgerson et al. explored dose response in neuroblastoma from 800 to 4000 cGy and found no difference in outcomes. For radioresponsive histologies, 600 to 900 cGy over 3 days has been recommended for urgent treatment. , Higher palliative doses of 2400 to 3000 cGy may be used for patients with soft tissue sarcomas and other resistant histologies. In a recent multi-institutional series the most common dose and fractionation schemes for spinal disease were 3000 cGy in 10 fractions, 2400 cGy in 6 fractions, and 750 cGy in 3 fractions ( Table 17.5 ).

TABLE 17.5
Most Frequent Radiation Therapy Regimens for Specific Anatomic Sites
Data from Rao AD, Chen Q, Ermoian RP, et al. Practice patterns of palliative radiation therapy in pediatric oncology patients in an international pediatric research consortium. Pediatr Blood Cancer . 2017;64(11).
Symptom Dose/# of fx
Bone
  • 2000/10

  • 800/1

  • 2000/5

Primary brain
  • 2400–3000/12

  • 3000/10

  • 3750/15

Brain metastases
    • WBRT

  • 3000/10

  • 200/5

  • 3000/15

    • SRS/partial

  • 3000/10

  • 3500/5

  • 5400/30

Spine
  • 3000/10

  • 2000/5

  • 750/3

Lungs
  • 2000/10

  • 2000/5

  • 300/4

Leukemia/lymphoma 2000/10
fx, Dose (cGy)/ # of fractions; SRS, stereotactic radiosurgery; WBRT , whole brain radiation therapy.

Brain metastases

In pediatric patients, brain metastases from a solid extracranial primary tumor is uncommon, with reported frequencies of 1.5% to 10% from clinical series and 6% to 13% from autopsy reports. In contrast to superior vena cava syndrome (SVCS) and cord compression, brain metastases are typically not the initial presentation of disease and most often occur after extensive systemic metastasis and disease progression. , , , Intracranial spread may arise from tumor in adjacent bony structures invading the meninges and brain parenchyma, more common in osteosarcoma or rhabdomyosarcoma, or by hematogenous spread from distant organs. There is a strong association of brain metastases following pulmonary metastases due to hematogenous spread. In a study from MD Anderson, 70% of patients had a lung metastasis at the time brain metastases were diagnosed.

The most common histologies include sarcomas, neuroblastoma, and germ cell tumors. Multiple series have suggested changes in epidemiology of pediatric brain metastases with the advent of modern chemotherapy regimens. Metastases from Wilms tumors were reported in up to 13% of patients dying of metastatic disease on autopsy but was clinically observed in no patients treated with modern systemic therapy. , , While brain metastases are historically rare in patients with neuroblastoma, prolonged survival with chemotherapy has resulted in increasing numbers of reports describing CNS relapse. ,

The most common presenting symptoms include headache, nausea, vomiting, and seizures. MRI is the most sensitive imaging modality for diagnosis of multiple lesions, although computed tomography (CT) may be obtained in the emergent setting. Neuroaxis spread has been reported in 26.7% to 36% of cases and cytologic evaluation with cerebral spinal fluid (CSF) is recommended to exclude or establish leptomeningeal disease (LMD) after imaging is obtained. , In general, upfront symptomatic management with steroids and anticonvulsives is indicated.

For patients with solitary brain metastases, controlled systemic disease, and good performance status, surgical management has been shown to be beneficial. Graus et al. reported that six patients with solitary brain metastases who underwent surgical resection followed by RT had a median survival of 7 months. In a report from Memorial Sloan Kettering, nine patients with neuroblastoma were treated with palliative radiation to doses from 1500 to 3000 cGy. The survival times for the four children treated with radiation alone ranged from one to five and a half months while survival for those treated with surgery followed by radiation ranged from 4.2 to 25 months. While craniotomy is typically offered only to select patients with good performance status and solitary lesions, these studies support use of surgical excision for single metastases when feasible.

Following surgical resection, postoperative RT is indicated. The importance was highlighted in a study by Bouffet et al., in which a patient treated with surgery alone had an early recurrence at the same site. Postoperative RT was subsequently recommended in all cases. Similarly, Paulino et al. found better freedom from neurologic progression in patients who received RT as compared to those who did not. In the postoperative setting in this series, patients were treated with whole brain RT to 2000 to 3000 cGy in 10 fractions.

Series reporting use of RT without surgical resection demonstrate a response based on histology. In a series of patients treated with whole brain RT alone to 3000 cGy in 10 fractions, a complete response was detected in four of five patients with germ cell tumors and only one of six patients with sarcoma. None of the patients with a complete response to radiation died of recurrent brain disease. For patients at initial diagnosis or with excellent prognosis, aggressive palliation with doses comparable to those used to control the primary tumor may be considered. Paulino et al. treated newly diagnosed metastatic Ewing and rhabdomyosarcoma with 2800 to 3000 cGy whole brain radiation followed by tumor boosts of 2800 to 5000 cGy. In contrast, Caussa et al. treated patients with a single metastasis from neuroblastoma to a median dose of 1500 cGy with a response rate of 44%. For patients with poor prognosis, doses as low as 900 cGy whole brain radiation may be considered for short-course palliation.

Long-term outcomes after treatment for brain metastases remain poor, with median survival reported from 4 to 9 months and 1 year survival rate from 11.5% to 38%. , , Suki et al. reported median survival of 0.9 months in patients who did not receive treatment after a diagnosis of brain metastases compared to 8 months in those who initiated therapy. In cases of intracranial metastases, deaths are related to both progression of intracranial and systemic disease. , However, deaths due to neurologic causes have been reported in only 20% to 27% of patients who receive RT. ,

Patients presenting with LMD from extracranial solid tumors have very poor outcomes, with median survival of 2 months reported in one series. For patients with poor performance status and uncontrolled extracranial disease supportive therapy alone is reasonable, while those with good clinical features may have improved survival with aggressive craniospinal irradiation (CSI). , Deutsch et al. reported on four patients with LMD from neuroendocrine carcinoma, squamous cell of the lung, retinoblastoma and rhabdomyosarcoma treated with CSI. Doses to the craniospinal axis ranged from 2200 to 3600 cGy. Patients survived between 4 days and 11 years. Given the limited number of patients and array of histologies associated with LMD, differentiation of outcomes between various therapies is limited. Optimal therapy must take into consideration the patient’s extent of extracranial disease, systemic therapy options, and patient and family preferences for aggressive treatment.

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here