Challenges facing survivors of childhood and adolescent cancer

In many pursuits, the ‘finishing line’ is not always where you think it is. For many years the ‘finishing line’ for children and adolescents with cancer was being told that they were cured – usually after four to five years of disease-free survival. It is now apparent that the consequences of having had a malignancy, especially in childhood, can impact many years later. In the greater scheme of things, curing significant numbers of childhood cancers has been a relatively recent phenomenon, namely over the last 35-40 years. This period has seen a phenomenal change in outcomes for childhood malignancies. When Farber first used methotrexate to treat children with leukaemia in 1948, short remissions resulted, but ultimately all patients succumbed to the disease. His initial report in the New England Journal of Medicine in 19481 was met with derision, as the prevailing view was that leukaemias were incurable and that the children should be allowed to ‘die in peace’. Indeed, in the 1960s parents were often told to take their children home and to love them, as there were no sustained remissions.

The use of multi-agent chemotherapy in the late 1960s led to the first reported durable remissions for children with acute leukaemias and lymphomas. The 1970s saw a dramatic rise in cure rates for many malignancies. It is humbling to realise that many of these children are now in their 40s and 50s - in many respects still relatively young.

Until this watershed, the only long-term survivors of childhood malignancy resulted from curative surgery or curative radiation. The numbers were small, but even then there was a recognised cost seen in these long-term survivors with growth effects, neuro-cognitive and neuro-endocrine complications and the suggestion of increased second malignancies. In his seminal paper on the role of radiotherapy in medulloblastoma, Bloom noted that children under two years of age often required ongoing institutional care after receiving craniospinal radiotherapy.2 Prior to this, Lampe expressed concern regarding brain damage that could result from high doses of radiation to brains of younger patients.3

It was hoped that chemotherapy would eliminate the need for radiation and be free of long-term consequences, but unfortunately this was not to be. Until the 1990s, once a patient was deemed cured they were usually discharged from the primary treating institution and told to live a normal life with reasonable expectation that they would. There has been an increasing recognition over the last 20 years of the many complications that may result from anti-cancer treatments. As a profession, we have an obligation to screen for and deal with these problems. This paper reviews the challenges facing childhood cancer survivors considering the physical, psychological, social and financial implications.

Epidemiology of cancer in children

According to the Surveillance, Epidemiology and End Results data, the incidence of cancers in young people less than 20 in the US, has remained static between 12.9 and 16.7 per 100,000 during the last 30 years.4 During this time, the mortality incidence has decreased from 5.2 to 2.6. The most common malignancies are leukaemias (45/100,000), central nervous system malignancies (28/100,000), lymphomas (24/100,000), soft tissue sarcomas (12/100,000), germcell/trophoblastic tumous/neoplasms of the gonads (11.6/100,000), malignant bone tumours (9/100,000), neuroblastomas (8/100,000) and renal tumours eg. Wilms (6/100,000).5

As a result of this improvement in treatment, it is now expected that 80% of childhood cancer patients will become long-term survivors.6 One in 640 young adults 20-39 are cancer survivors and this means that the average general practice would be expected to have at least two of these patients per physician.

Treatments for childhood and adolescent cancers are diverse. The most common malignancy in this group is leukaemia (as above) and the chemotherapy for this condition can often continue over 18 months, possibly requiring total body irradiation and bone marrow transplantation. Lymphomas are treated predominantly with chemotherapy, with radiation used in a number of cooperative group protocols, albeit at much lower doses than adult patients receive. Brain tumours are usually treated with up-front surgery followed by up to 59 Gy of radiation. Sarcomas are usually treated with surgery and subsequent chemotherapy, often with the role of radiotherapy dependant on the histology and surgical clearance pathologically. The doses of radiotherapy for this group are variable, ranging from a modest 36 Gy to a radical 50.4 (Rhabdomyosarcoma) or 55.8 Gy (Ewings). At the other end of the spectrum, Wilm’s tumour in the early stages is treated with a short course of chemotherapy and radiotherapy usually of very small doses (10.8-21.6 Gy).

The gist of the European cooperative group studies has been to avoid radiation unless the local control with surgery comes with unacceptable toxicities. The North American cooperative groups in contrast have aimed to reduce the doses of radiotherapy, but the result is that more children are receiving it. Not surprisingly, over the last decade many treatment protocols for diseases such as Wilms tumour are converging on the role and dose of radiotherapy, allowing cross group studies to be entertained.

Physical effects from the cancer itself

The tumour itself can cause significant long-lasting problems prior to any therapeutic intervention. In brain tumours, there is good evidence that having a tumour itself can cause disturbance of the hypothalamic pituitary axis prior to surgery or radiotherapy.7 Likewise, the development of hydrocephalus is recognised to be an independent cause of significant neurocognitive decline in patients with brain tumours, previously attributed solely to radiation therapy.8 Damage to neurones may not be repairable and so often timely intervention is crucial in the setting of cord compression (eg. osteosarcoma or Ewing’s), or the optic chiasm (craniopharyngiomas and optic pathway gliomas).

Surgery

Clearly, the need for cancer resections of bony structures may have significant cosmetic effects and impacts on growth, especially if the growth plates are involved. The issue of rehabilitation involved after amputations is significant. Those patients who have undergone splenectomy, as performed at staging laparotomy for Hodgkin’s disease in the 1970s and 80s, are at risk of pneumococcal and meningococcal infections and as such require life-long surveillance and vaccinations.9 Long-term neurocognitive insult and neuronal injury are a possible complication of major brain tumour resections. Nephrectomy patients develop compensatory hypertrophy of the remaining kidney and run the risk of earlier onset hypertension, with its related health issues and proteinuria.10-12

Late effects from radiotherapy and chemotherapy

The most famous first victim of radiation late effects was probably Marie Curie, who discovered radium along with her husband Pierre. She also went on to develop the first mobile X-ray station in France in World War I. Marie died of aplastic anaemia, most likely as a result of her long-term radiation exposure. Her daughter Irene, also a Nobel Prize winning radiation physicist, developed and died from acute leukaemia. Pierre Curie however, was spared a similar fate - he was run over by a horse drawn cart on the streets of Paris in April of 1906.

The most studied modality producing late effects is radiation. The first patients were treated in the late 1890s and until the advent of chemotherapy, it was the only effective non-surgical treatment for cancer available. However, from relatively early on the effects of radiotherapy were appreciated.

 “The dangers from the use of X-rays may be grouped as immediate and remote. During the actual exposure, the possibility of making contact with a high-tension lead carrying a very high voltage has to be guarded against. An accident of this kind may easily be fatal…constitutional disorders, anaemia and sterility not infrequently arise in operators who are constantly exposed to X-rays.”13

In 1935, the concept of immediate and long-term or late effects was very simple. It is now thought that late effects refer to complications that arise many months to many years after the completion of therapy.

Indeed, much of the significant early data regarding adverse effects from radiotherapy is not from therapeutic radiation exposure - rather from the Hiroshima and Nagasaki atomic bomb data, industrial accidents and use in benign conditions. For example, in the 1940s and 50s, superficial irradiation was a commonly used treatment for tinea capitis, with doses of 0.04-0.45 Gy used.14 Early reports from the 1960s suggested an increase in leukaemias, other malignancies and interestingly ‘mental disorders’. The incidence of thyroid, brain and other head and neck cancers was also found to be increased in the large cohort of Israeli immigrants treated for tinea in the 1940s and 50s.15

Much of the current data regarding late effects of cancer treatments has been developed for the retrospective cohort of ~ 10,000 patients with matched sibling controls in the Childhood Cancer Survivors Study group.6,16-19 Much of this data and other published literature has been brought together in the formation of the long-term follow-up guidelines of the Children’s Oncology Group (www.survivorshipguidelines.org/). These guidelines are used as the basis for many long-term follow-up programs both in the US and internationally.

It is beyond the scope of this paper to exhaustively detail the physical effects of chemotherapy and radiotherapy, however, a brief overview follows.

Head and neck region

Alopecia is physically perhaps the most insignificant side-effect of cancer treatment, but psychosocially, one of the more distressing, particularly for teenage girls. Cranial radiation often leads to temporary hair loss in field and the degree of permanent effect relates to total dose. It is apparent that combined chemoradiation (such as in medulloblastoma) can lead to more pronounced permanent thinning of the hair, or indeed complete alopecia.20

The lens of the eye is very sensitive to the effects of radiation and to steroid administration, which both lead to cataractogenesis.21 In conditions such as medulloblastoma and leukaemia requiring prophylactic cranial irradiation, it is difficult to adequately cover the cribriform plate without giving some dose to the lens. In itself, cataract removal is a fairly straightforward procedure, but the dose to the anterior chamber of the eye also increases the later risk of developing glaucoma.22

The hypothalamic - pituitary axis is often compromised if it is involved in the surgical resection of tumours (particularly craniopharyngioma). Both surgery and radiotherapy to the hypothalamus can lead to hypothalamic obesity or metabolic syndrome, which is thought to be due to an abnormality in the normal satiety response to food. Radiotherapy effects to this axis present with a median time of three years post therapy. The thyroid axis is usually affected first, followed by growth hormone, the sex hormones (sometimes presenting as  precocious puberty) and less commonly Adrenocorticotropic hormone, leading to Addisonian syndromes. The thyroid gland itself may suffer primary failure if it is in the primary radiation field. In conditions requiring cranio-spinal irradiation, it may prove difficult to distinguish between central failure and peripheral (glandular) failure. Central infertility may also result from radiation, however, this may be negated by the use of gonadotrophic releasing hormone agonists to induce gonadal stimulation.23-26

Often the most devastating long-term effects is the functional neurological compromise suffered by patients who have had brain tumours or cranial irradiation.27 As mentioned previously, there is evidence that hydrocephalus itself can aggravate neurocognitive compromise.8 Merchant et al have demonstrated that IQ decline is proportional to the volume of brain treated, especially the temporal lobes and the dose these volumes received.28 Palmer et al found that there appears to be a constant decline until age 12, after which the IQ remains stable. There is however, a progressive reduction in short-term memory and concentration span through the teenage years.29 Some evidence suggests medications such as dexamphetamine and/or cognitive remediation programs may improve academic performance and overall quality of life in some with a history of brain tumours.23,24,30,31 Similar, but not as profound effects can be seen in patients who have had intrathecal methotrexate, especially if cranial radiotherapy is also given.25,32

Radiation can age the brain and there is a small risk of focal radionecrosis in high dose regions,33 as well as a general increased risk of cerebrovascular accidents. Radiation to the neck and mediastinum can increase the rates of cerebrovascular disease in five-year survivors from Hodgkin’s disease.34 For this reason, many late effects services take an aggressive approach to management of hypercholesterolemia, hypertension and other reversible risk factors for cerebrovascular disease.

Cardiac effects

Both radiotherapy and chemotherapy have significant impacts on cardiac function. High-dose anthracyclines (eg. > 350 mg/m2 equivalent doxorubicin), can induce cardiac failure during treatment.35 There is also a recognised decrement in cardiac function which may present years later.36,37 In female patients, cardiac failure may be unmasked during pregnancy. It is advisable for pregnant women with a history of cardiac irradiation or anthracycline chemotherapy to undergo cardiac function assessment during pregnancy and monitoring during labour and delivery. Radiotherapy to the chest increases the risk of ischaemic heart disease by 2-5%.19,38 These patients also have an increased rate of valvular abnormalities, usually presenting with stenotic rather than incompetent valvular heart disease. Renal irradiation may cause cortical scarring or fibrosis, increasing the risk of Angiotensin converting enzyme driven hypertension, aggravating both the cerebral and cardiac risk profile.10

Intriguingly, there is data that implicates higher doses of cisplatin used in the treatment of testicular cancers in the development of the metabolic syndrome. This clearly has ongoing implications for the cardiovascular health of these patients.39

Pulmonary effects

Radiation doses above about 20 Gy induce variable degrees of pulmonary fibrosis in the radiation field, which if marked, may lead to a restrictive pattern on lung function testing and a decrease in overall diffusing capacity.32,40 This is particularly relevant in patients with mediastinal lymphoma, and needs to be considered in some with a history of neuroblastoma. Bleomycin chemotherapy is a potent inducer of interstitial fibrosis and pneumonitis.33,41 These problems are aggravated by tobacco and marijuana smoking, so smoking cessation is essential for people with these prior exposures.

Gastrointestinal effects

High dose radiation to the gastrointestinal tract can lead to localised strictures, gastrointestinal tract blood loss from telangectatic blood vessel formation in the walls and/or chronic loose motions or diarrhoea.1,42 There are reports of radiation induced bowel cancers in people treated for Wilm’s tumour or rhabdomyosarcoma.43

Genitourinary effects

High dose cyclophosphamide may induce haemorrhagic cystitis despite routine intravenous hydration prior to chemotherapy. Cyclophosphamide in childhood increases the risk of later bladder malignancies.44 High dose irradiation may induce scarring in the bladder, causing reduced bladder volume, which may result in urinary frequency and urge incontinence. Many chemotherapies, particularly potent alkylating agents (especially nitrogen mustard), can induce infertility in later life.45 Chemotherapy (especially cyclophosphamide) may be associated with premature menopause. This is related to chemotherapy dose and age at treatment. Radiation has been found to reduce uterine blood flow, and in doses above 16-20 Gy may induce hypoplasia and fibrosis, resulting in miscarriage or inability to carry a pregnancy to term. Radiotherapy doses of 2-4 Gy to the testes and 4-6 Gy to the ovaries may induce sterility, and at higher levels (~20 Gy) may result in loss of hormonal function.46-48 

Musculoskeletal hypoplasia

As depicted in figure 1, the threshold dose for hypoplasia induced by radiation appears to be about 16 Gy, with the plateauing of effect seen at about 25 Gy. If there is inhomogeneity across growth plates (as in vertebral bodies), asymmetric growth may lead to impaired cosmetic outcomes, such as kyphoscoliosis, facial asymmetry and pelvic tilt. Associated with this could be effects on neuronal, glandular and mechanical functions as described above. Clearly another mechanism of impaired growth is from the effects on growth hormone production from hypothalamic/pituitary irradiation. Chemotherapy itself may cause overall growth failure, with twin studies showing that bone marrow transplanted patients are reliably shorter than their siblings. Radiation can lead to late osteoporosis in field and in some cases radionecrosis in high dose areas. Likewise, high total dose corticosteroids may induce osteoporosis and more worryingly avascular necrosis of the head of the femurs.