Malignant Neoplastic Diseases Diagnosed in the Infancy Period
The spectrum and clinical aspects of malignant neoplastic diseases diagnosed in the infancy period are quite distinct from those found in older children. However, there are very few published data focusing on this particular group of patients. We performed a retrospective study on our paediatric cancer patients in order to review the spectrum and clinical aspects of infant malignant diseases. A total of 40 infants with various malignant neoplastic illnesses and Langerhans cell histiocytosis (LCH) were diagnosed in our unit within a 10 years period (Jan 89 to Dec 98). It accounts for 13.5% (40/296) of the total paediatric oncology patients diagnosed within the same time span. Blastoma is the most common group of illnesses (n=19) in the infancy period, followed by leukaemia (n=8) and brain tumour (n=7). Malignant blastoma in the infancy period appears to have good outcome. But infants with acute leukaemia and brain tumour had poorer prognosis as compared to older children with the same diseases. Even within the same disease category, the biological behaviour of infant neoplasms appears to be different from that of the older children.
Keyword : Children cancer; Infant
The incidence, spectrum, natural history, clinical aspects and response to treatment of childhood malignant diseases had been shown to be different from that of the adult. For example, more than 90% of the adult cancer are carcinoma arising from different organs.1 In contrast, carcinoma only accounts for less than 3% of paediatric malignant tumours (unpublished data Hong Kong Paediatric Haematology / Oncology Study Group 1981 to 1992). In addition, children with cancer have relatively better outcome than adults with cancer. Approximately 50 to 60% of all childhood cancer can be cured.2 Whether very young children with cancer differ from the older children with malignant diseases has not been properly looked into. In several disease entities, certain pattern of differences has been demonstrated. For example, infants with acute lymphoblastic leukaemia have a distinct biological nature and poorer outcome,3 infants with neuroblastoma may regress completely despite clinical evidence of metastasis,4 infants with disseminated Langerhans cell histiocytosis (LCH) tend to have aggressive clinical behaviour and poor prognosis.5 We also noted that several diseases such as retinoblastoma, hepatoblastoma tend to occur early in life. These observations are urging us to look at the whole age group from a more holistic view. Our objectives are 1) to find out the relative incidences of various malignant diseases found in the infancy period; 2) to study their presenting pattern, clinical outcome and response to therapy; 3) to identify any clinically significant therapy-related adverse event in this vulnerable age group.
We performed a retrospective records review on our paediatric cancer patients from January 89 to December 98. Infant is defined as a child less than or equal to one year of age. Only children with malignant neoplastic diseases or LCI-I were included in this study. LCH was included because its infantile disseminated form can affect multiple organs and carry a very poor prognosis. Benign tumours such as cystic hygroma, sacrococcygeal teratoma and haemangioma were not included. This group of benign tumours were excluded in this review because they are often not referred to the paediatric oncology service and the actual incidence will therefore tend to be underestimated. Transient abnormal myelopoiesis (leukaemoid reaction) associated with Down syndrome was also excluded because it is often transient in nature and requires no intervention. Since the data are extracted from a single tertiary referral centre, some bias in the disease distribution is expected.
The results were analyzed by the computer statistical software system (SAS). Survival analysis was by Kaplan-Meier method and paired-wise comparison of outcome of the different disease groups were made by Chi-squared test and log-rank analysis.
Of the 40 infants with malignant neoplastic diseases, 23 of them were male. The median age at diagnosis was 8-month-old (range: at birth to 12 months). Only 3 infants (retinoblastoma n=1, primitive neuroectodermal tumour n=1, malignant germ cell tumour n=1) were diagnosed to have the disease during the neonatal period which is defined as within the first month of life. Thirteen patients had their first presentation within the first 6 months of life and leukaemia (n=5), hepatoblastoma (n=3) and retinoblastoma (n=2) were the most common groups of cancer found in this period.
Concerning the disease type (Table), blastoma (n=19) which included a heterogenous group of primitive embryonal tumours arising from various organs was the most common group of tumours found in the infancy period. The blastomas include peripheral germ cell tumour, hepatoblastoma, retinoblastoma, neuroblastoma and nephroblastoma (Wilm's tumour). Some of these blastomas, such as hepatoblastoma and retinoblastoma, were relatively uncommon in the older children but they were over-represented in the infant population. In contrast, some relatively common solid tumours found in the older children such as lymphoma, osteosarcoma were not found in our infant population. The comparison of the relative proportion of various tumours in general paediatric population based on the HKPHOSG (Hong Kong Paediatric Haematology Oncology Study Group) data and our infant cancer patients were shown in fig. 1.
Acute leukaemia was the second common group of infant malignant neoplastic diseases found in our unit [acute lymphoblastic leukaemia (ALL) n=6, acute myeloid leukaemia (AML) n=2]. It was followed by brain tumour. In contrast to the older age group of children in which medulloblastoma accounts for 25% of the cases of brain tumour, astrocytoma (n=2); ependymoma (n=2) ; cerebral primitive neuroectodermal tumour (n=l); and some rarer forms of brain tumour (ependymoblastoma n=1, choroid plexus papilloma n=1) were encountered in almost equal proportion in our infant cohort. In a strict sense, choroid plexus papilloma should be considered as a benign rather than malignant brain tumour.
In general, the overall survival of our infants with malignancy was 73% (median follow-up 3 years) (Fig. 2). The apparently better prognosis was mainly contributed by the higher incidence and better outcome of the group of blastomas in infancy. They were very responsive to conventional chemotherapy and had an excellent outcome. In our series of infant blastomas, the 3 year overall survival was 95% and event free survival was 89%. The only casulty was a patient with stage 3 neuroblastoma who died of blood transfusion-related graft versus host disease.
We did not find any stage 4S neuroblastoma in this ten-year period. One child with a stage 2 neuroblastoma probably has congenital neuroblastoma for she was treated for pneumonia at the age of one month. The chest x-ray at that time showed left upper lobe "consolidation" and it was later confirmed to be localized neuroblastoma. We excluded 4 cases of benign sacrococcygeal teratoma from our review. For the remaining 7 patients with malignant peripheral germ cell tumour, 6 of them had yolk sac tumour and one had malignant teratoma. Retinoblastoma was identified in three patients (unilateral n=2, bilateral n=1), one was detected incidentally by the ophthalmologist during the routine examination for retinopathy of prematurity. The other two were picked up by incidental detection of "white eye reflex". All of them had localized diseases and two of them responded to conservative therapy (laser or localized radiation therapy). One had to undergo enucleation of the affected eye ball due to more extensive involvement. All of our five patients with hepatoblastoma survived and it is alarming to note that 3 of the 5 patients suffered from either spontaneous rupture of the tumour or tumour vasculature while waiting for definitive therapy within the hospital. They were all subsequently salvaged but this highlighted the importance of early referral and treatment of this group of patients. Both of our infant-onset nephroblastoma (Wilm's tumour) patients had localized disease and were long term survivors. None of our infants with blastomas had any underlying syndromal condition.
Our infants with acute leukaemia had an overall 3 year overall survival of 56% and event free survival of 44%. If we look at the infants with ALL alone, the 3 year overall survival was 50% but event free survival was only 16%. This is in line with the published data that infants with ALL performed much poorer than the older age group with the same disease (overall 3 year event free survival 70%).6 The seemingly better survival rate of our infants with acute myeloid leukaemia (overall survival 100%, median follow-up 2 years) as compared to the experience in older children was probably due to the early implementation of bone marrow transplantation (BMT).7 In fact, three of our 4 infants with leukaemia who underwent BMT were long term survivors and both of the infants with AML underwent BMT and survived.
The 3 year overall survival of infants with brain tumours was 42% which is slightly worse than that of the older children (overall survival 55%).8 That may be related to the smaller number of chemo-sensitive tumour type (e.g. medulloblastoma) identified in this age group. Besides, cranial irradiation will not be given as a primary form of treatment in patients less than three years of age due to its potential neurological damage on the developing brain. Upfront chemotherapy is the only alternative in most setting.
Disseminated Langerhans cell histiocytosis (LCH or previously known as histiocytosis X) in infancy is a very aggressive and highly fatal disease. Of our 4 infants with LCH, all had multi-organs involvement (Lavin's Stage IV)9 and 2 died of disease progression despite chemotherapy. One was lost to follow-up (patient returned to Indonesia). One responded well to chemotherapy initially but relapsed around 1.5 year later. She entered into good remission again after another course of chemotherapy. We have another 3 patients with LCH who were diagnosed in their second year of life, they also had multi-organs involvement (Lavin's stage II, n=1, stage III, n=2) but all survived with chemotherapy. Two of them experienced the typical wax and wane clinical courses of LCH and required multiple courses of chemotherapy.
For soft tissue sarcoma, the only cases that we had were two infants with localized peripheral primitive neuroectodermal tumour (PNET). Both infants had very aggressive disease and did not respond to chemotherapy. More common soft tissue sarcomas such as rhabdomyosarcoma were only found in our older patients cohort. This may be due to our small infant cohort sample size.
By log-rank analysis, we found that infants with blastomas had significantly better overall and event free survival than infants with acute leukaemia and brain tumour (overall survival, p=0.034; disease free survival, p=0.047) (Fig. 3). We did not identify any of our patients with clinically significant long term therapy-related cardiac, hepatic or renal dysfunction. One of our infants with ALL who underwent BMT with local testicular irradiation experienced primary growth failure. He did not have any evidence of central nervous system disease and did not receive cranial irradiation. Whether the primary growth failure was related to the intrathecal methotrexate given as central nervous system prophylaxis remains unknown.
Our data showed that the spectrum and clinical aspects of malignant neoplastic diseases diagnosed in the infancy period are quite distinct from those found in older children. Blastoma is the predominant group of tumours found in this age group and they have excellent prognosis if treated properly. Acute leukaemia is another group of commonly encountered malignancy but unlike the older children, its prognosis with the conventional therapeutic approach is very disappointing. Another important group of disease is LCH, not only that it is more commonly found in the infancy period, it also has a much more aggressive clinical course in this age group.
Among the blastomas, neuroblastoma is an interesting one. Its initial presentation depends on its location, and the classical description is a toxic, pale and irritable looking infant with a distended abdomen. The liver may be enlarged, especially in infants with stage 4S disease. If it is located in the thoracic region, it can present with a supraclavicular mass associated with Homer's syndrome. An elevated urine HVA and VMA level is a good screening test but up to 5% of the neuroblstoma patients may not have elevated urinary catecholamine.10 While the prognosis of metastatic neuroblastoma remains poor with the current available treatments,10,11 many of the metastatic neuroblastoma (stage 4S) found in the infancy period may regress spontaneously.12 The recent advances in the understanding of tumour biology can explain partly of this peculiar phenomenon. The stage 4S neuroblastoma does not have the poor prognostic features such as Myc amplification, chromosome 1p deletion. Instead it carries some favourable features such as hyperdiploid DNA index, low serum ferritin level.13 Recent study has also shown that stage 4S neuroblastoma does not express telomerase14 and has a shorter telomeric length,15 a sign of lesser immortalized potential than other aggressive cancer. These differences in biological nature are also found in the stage 1 and stage 2 neuroblastoma. Most paediatric oncologists nowadays would consider stage 1, 2 and 4S neuroblastoma as separate entities when comparing with the stage 4 neuroblastoma. There is so far no evidence to suggest that low stage neuroblastoma will progress into advanced stage neuroblastoma. The concept of early detection of neuroblastoma by mass screening to prevent early stage disease from progressing into advanced stage disease is therefore theoretically unsound. That is also the reason why screening of neuroblstoma was not found to be effective because it mainly picked up the early stages neuroblastoma and the artefactual improvement of the survival may only represent the good survival of the early stage disease.16 As we know, the early stage neuroblstoma can either spontaneously regress or differentiate into mature tumour, aggressive therapy may not be necessary.
Peripheral germ cell tumour is relatively more common in the infancy period. The benign form such as sacrococcygeal teratoma may present as a buttock mass at birth. The malignant form may present as a buttock mass or pelvic mass with associated neurological deficit such as urinary retention and lower limb weakness, or as a testicular mass. Serum markers in the form of αFP (yolk sac tumour), βHCG (choriocarcinoma) can help us to further differentiate the subtypes of germ cell tumour. Malignant teratoma and dysgeminoma do not secret any serum marker and the mixed type may secret more than one form of serum markers. Except for the malignant teratoma, most types of germ cell tumour are chemo and radiation therapy sensitive. In infant, chemotherapy should be the main treatment modality beside surgery. The therapy-related toxicity of chemotherapy is manageable with proper dosage adjustment. In our series, only one of the seven infants suffered from recurrent disease after chemotherapy and he was rescued by another regimen of chemotherapy. He has been off treatment for 8 years now without major complication. Another important point for sacrococcygeal germ cell tumour is that the coccyx should be removed in the initial surge to prevent possible disease recurrence.17
For infants with hepatoblastoma, they often present with abdominal distension and cholestatic jaundice is a late feature. The serum αFP level is usually very high in the range of 106 ng/ml level which is way beyond the maximal age-related limit in the first month of life (usually <103 ng/ml). Unlike hepatocellular carcinoma, hepatoblastoma has no correlation with the maternal hepatitis status. Though not reported as a common feature previously, we noticed that spontaneous rupture of the tumour and tumour related vessels was not an uncommon event. It can be fatal if emergency operation cannot be implemented on time. Three of our five patients experienced this catastrophic event but they were all saved by emergency operation. Hepatoblastoma is a chemo-sensitive tumour and our patients responded very well with the current regimen.18 Only one patient has persistent mildly elevated serum αFP level but no overt clinically detectable tumour. This is suggestive of residual tumour, he is currently undergoing autologous bone marrow transplantation.
Due to the pattern of referral, we do not have many retinoblastoma patients in our cohort. The early detection of "cat eye or white eye reflex" is important. Two of our patients were found by their parents to have white instead of red eye light reflection to the flash light in the picture, they were then referred and confirmed to have retinoblastoma. With early detection, local eye-saving treatment (laser or local irradiation) may be possible.19 Except for metastatic retinoblastoma, chemotherapy is usually not indicated.
Nephroblastoma or Wilm' s tumour is a disease of young children. Again with early detection, the prognosis is excellent.20 Most of the patients presented with an abdominal mass incidentally detected by the patient's family. They are usually asymptomatic and haematuria or hypertension is uncommon. The current treatment approach is surgical excision followed by a stage-directed chemotherapy regimen. The treatment for the early stag disease is quite short and relatively non-toxic.21 But precaution has to be exercised, not all renal tumours in children are nephroblastoma. For example, mesoblastic nephroma can be found in infant and surgery is the only form of therapy needed if it can be excised.22 Other rarer renal tumours such as clear cell sarcoma, renal primitive neuroectodermal tumour and rhabdoid tumour are usually found in the older children and they require more aggressive therapy.21
Leukaemia in infancy shares the similar presentation as older children. Persistent fever, pallor, multiple petechiae and bruises in an infant with poor appetite is suggestive of serious infection, marrow failure or leukaemia. The presence of lymphadenopathy and hepatosplenomegaly would further support this suspicion. Unlike the older children, infants with acute lymphoblastic leukaemia (ALL) have been shown to have extremely poor prognosis if treated with chemotherapy alone. That is because a large proportion of the infant ALL had t(4; 11), a translocation known to be associated with poor treatment outcome,23 whereas only less than 3% of the older children have this translocation.(Look T. Unpublished data) In addition, t(12;21), a favourable translocation commonly found in 25% of children with ALL, is uncommon in infant.24 Our infant ALL patients showed a similar picture of poor outcome as described in the literature except that t(4;11) was only identified in one of the 6 ALL infants. A detailed description of our infant ALL patients has been reported previously.25 A more sensitive way to detect this translocation by molecular method (RT-PCR multiplex) has been established in our unit since 3 years ago and this may be able to avoid under-diagnosis of this genetic event.26 Interestingly, both of our infant AML patients also had translocation involving the MLL gene in chromosome 11q23, the same gene involved in the t(4;11) in infant ALL. The t(11;v) in childhood AML is also associated with a grave prognosis,27 but both of our patients survived after they underwent allogeneic bone marrow transplantation.
The relative frequency of brain tumours found in infancy period is not the same as in the older children. Based on the study by Gerlach et al.,28 the most common types within the first year of life are choroid plexus papilloma or carcinoma (19%), central nervous system germ cell tumour (14%) and various form of astrocytomas (22%). Medulloblastoma only accounts for less than 10% of the cases. Due to the small sample size, we did not find any special pattern in our patient cohort and infratentorial tumour does not appear to be as common in this age group as in the older children. Only 2 of our 7 patients (28.5%) had infratentorial tumour in the form of ependymoma. Even with different tumour types, the presenting symptoms are quite similar in most cases. They usually had rapidly increasing head size or an incidental finding of a big head with bulging fontanelles as their initial presenting feature. Irritability and vomiting may occur but usually signify a late detection. Convulsion is uncommon (one of seven in our cohort). The treatment option is mainly surgery with or without chemotherapy. We usually withhold radiation therapy till 3 years of age if possible because of the severe neurological damage induced by this form of therapy in a developing brain. The central nervous system primitive neuroectodermal tumour, medulloblastoma and germ cell tumour usually responded to chemotherapy quite well.29 There is a recent report suggesting a short course of chemotherapy followed by megatherapy may provide good treatment outcome with lesser neurodevelopmental hindrance.30 This approach requires further study verification. The infancy ependymoma usually had poor prognosis and has to be followed up closely. A peculiar form of astrocytoma that is found in the infancy is chiasmatic pilocytic astrocytoma. Though it is a low grade tumour, surgical resection is often impossible due to its location and the patient is usually blind. W. have two patients with this form of astrocytoma. Despite the initial response, one patient's family decided to withhold treatment and the patient died around one year later. Whereas for another patient who went through 12 courses of chemotherapy, the tumour has remained static and he has been off treatment for more than 2 years now.
The disseminated LCH found in the infancy is a very aggressive form of LCH. Unfortunately, it is usually diagnosed late in affected infants because the eczematous skin and scalp lesions, otitis externa were frequently mistaken as usual infants problem and was treated as a localized skin disease. All of our infants with disseminated LCH had a delay in diagnosis by several months. The refractory nature of the skin lesions and ear discharge together with the early onset of gum swelling should alert a paediatrician about the possibility of LCH. The diagnosis is usually confirmed by skin biopsy and the classical "Birbeck granules" demonstrated by electron microscopy. Although there is still controversy about the actual nature of LCH, it is not considered as a form of malignancy by most experts. On the other hand, there is not much argument about the need of using chemotherapy for disseminated form of LCH. As of now, there is no uniform agreed chemotherapy regimens but the trend is to treat the disease aggressively upfront. We adopted the low dose cytarabine (Ara-C) and steroid regimen recently and the initial result has been encouraging.31 The family has to be cautioned that multi-organ LCH tends to recur and organ dysfunction such as diabetes insipidus, hypopituitarism may appear months or years after the initial presentation.
It is surprising to note that soft tissue sarcoma is quite uncommon in our infant patient cohort. Both of our patients suffered from peripheral PNET. They presented as soft tissue mass arising from their trunk or back. In contrast to our experience with PNET in the older age group,32 both of them did not respond to chemotherapy or radiation therapy and the detail of one of these 2 patients has been reported.33 Other soft tissue sarcoma such as rhabdomyosarcoma, especially the typical sarcoma botryoides arising from the vagina of female infant, has not been encountered by us in this ten year review. The rhabdomyosarcoma of the infancy is usually of the embryonal type which has a better prognosis.34
In summary, malignant diseases in the infancy period have an unique spectrum and clinical aspects as compared to the older children. To avoid missing the diagnosis, awareness of their relative incidence and usual pattern of presentation will be helpful. Due to the immature physical developmental state and organ functions, the dosage of medications and the choice of therapy often has to be modified. If managed appropriately, the majority of the infants with cancer can be cured. Our results compare favourably with those reported from major oncology centres in developed countries. Long term follow-up is necessary to monitor the potential therapy-related toxicity which may emerge many years later.
I would like to thank the medical and nursing staff of the Departments of Paediatrics and Surgery, Queen Mary Hospital for providing patient care, the members of the scientific committee of the Hong Kong Paediatric Haematology Oncology Study Group for providing local data and setting-up standard treatment protocols, Mr. Wilfred H.S. Wong for the statistical analysis and Ms. Olive Yu for the secretarial support.
1. Hong Kong Cancer Registry. Cancer incidence and mortality in Hong Kong: 1993-1994. Hospital Authority Jan.1998.
2. Schwartz CL, Hobbie WL, Constine LS. Overview. In: Schwartz CL, Hobbie WL, Constine LS, Ruccione KS, editors. Survivors of Childhood Cancer : Assessment and Management. Mosby, St. Louis, 1994:1-6.
3. Chessells JM, Eden OB, Bailey CC, Lilleyman JS, Richards SM. Acute lymphoblastic leukemia in infancy: experience in MRC UKALL trials. Reports from the Medical Research Council Working Party on Childhood Leukaemia. Leukemia 1994;8:1275-9.
4. Carlsen NL. Neuroblastoma: epidemiology and pattern of regression. Problems in interpreting results of mass screening. Am J Pediatr Hematol Oncol 1992;14:103-10.
5. Gadner H, Heitger A, Grois N, Gatterer-Menz I, Ladisch S. Treatment strategy for disseminated Langerhans Cell Histiocytosis. Med Pediatr Oncol 1994;23:72-80.
6. Ma SK, Chan GCF, Ha SY, Chiu DCK, Lau YL, Chan LC. Clinical presentation, hematologic features and treatment outcome of childhood acute lymphoblastic leukemia : a review of 73 cases in Hong Kong. Heamatol Oncol 1997;15:141-9.
7. Ha SY, Li CK, Lee CW, Yuen HL, Ling SC, Yuen PMP, Shing MK, Chik CW, Chan CF, Lau YL, Luk CW, Wong KW, Li CK. Childhood acute myeloid leukaemia in Hong Kong. H K Med J 1998;4(4 Suppl):61.
8. Heideman RL, Packer RJ, Albright LA, Freeman CR, Rorke LB. Tumors of the central nervous system. in: Pizzo PA, Poplack DG, editors. Principles and practice of pediatric oncology. 3rd ed. Lippincott-Raven, New York, 1997;633-97.
9. Lavin PT, Osband ME. Evaluating the role of therapy in Histiocytosis-X. Hematol/Oncol Clinics of North Am 1987;1:35-47.
10. Chan GCF, Shing MK, Luk CW, et al. Childhood neuroblastoma a review of the local experience. HK Med J 1998;4(4 Suppl):59.
11. Hartmann O, Valteau-Couanet D, Benhamou F, Vassal G, Rubie H, Beaujean F, Lemerle J. Stage IV neuroblastoma in patients over 1 year of age at diagnosis: consolidation of poor responders with combined busulfan, cyclophosphamide and melphalan followed by in vitro mafosfamide-purged autologous bone marrow transplantation. Eur J Cancer 1997;33:2126-9.
12. Miale TD, Kirpekar K. Neuroblastoma stage IVS. Med Oncol 1994;11:89-100.
13. Grosfeld JL, Rescorla FJ, West KW, Goldman J. Neuroblastoma in the first year of life: clinical and biologic factors influencing outcome. Semin Pediatr Surg 1993;2:37-46.
14. Hiyama E, Hiyama K, Yokoyama T, Matsuura Y, Piatyszek MA, Shay JW. Correlating telomerase activity levels with human neuroblastoma outcomes. Nat Med 1995;1:249-55.
15. Hiyama E, Hiyama K, Yokohama T, Ichikawa T, Matsuura Y. Length of telomeric repeats in neuroblastoma: correlation with prognosis and other biologic characteristics. Jpn J Cancer Res 1992;83:159-64.
16. Carlsen NL. Neuroblastoma: epidemiology and pattern of regression. Problems in interpreting results of mass screening. Am J Pediatr Hematol Oncol 1992;14:103-10.
17. Lahdenne P, Heikinheimo M, Nikkanen V, Klemi P, Shiner MA, Rapola J. Neonatal benign sacrococcygeal teratoma may recur in adults and give rise to malignancy. Cancer 1993;72:3727-31.
18. Chan KL, Saing H, Fan ST, Ng I, Chan CT, Ha SY, Lau YL. Primary paediatric liver tumours: Queen Mary Hospital experience. HK J Paediatr 1996;1;60-3.
19. Shields JA, Shields CL, Sivalingam V. Decreasing frequency of enucleation in patients with retinoblastoma. Am J Ophthalmol 1989;108:185-8.
20. Shing MK, Chan GCF, Luk CW, et al. Treatment of Wilm's tumour in Hong Kong. HK Med J 1998;4(4)(Suppl):60.
21. Green DM, Norkool P, Breslow NE, Finklestein JZ, D'Angio GJ. Severe hepatic toxicity after treatment with vincristine and dactinomycin using single-dose or divided dose schedules: a report from the National Wilm s Tumor Study. J Chin Oncol 1990;8:1525-30.
22. Kahidasan V, Mammen A, Hutson JM, Kelly JH, Auldist AW. Congenital mesoblastic nephroma: treatment options. Pediatr Sung Int 1994;9:524-5.
23. Pui CH, Ribeiro RC, Campana D, Ramondi SC, Hancock ML, Behm FG, Sandhund JT, Rivera GK, Evans WE, Crist WM, Krance R. Prognostic factors in the acute lymphoid and myeloid leukemias of infants. Leukemia 1996;10:952-6.
24. Shurtloff S, Buijo A, Behm FG, et al. TEL/AML1 fusion resulting from a cryptic t(12;21) is the most common genetic lesion in paediatric ALL and defines a subgroup of patients with an excellent prognosis. Leukemia 1995:9;1985-9.
25. Ma SK, Ha SY, Chan GCF, Chan LC, Lau YE. Acute leukaemia in infants less than 18 months of age: a retrospective analysis of 13 cases. HK J Paediatr 1996;1;37-43.
26. Chan GCV, Chart SY, Lee D, et al. Detection of genetic defect of childhood acute leukaemia by reverse transcriptase polymerase chain reaction (RT-PCR) : An essential component for risk directed therapy approach. Annual Scientific Meeting of the Paediatric Society of Hong Kong. Hong Kong, Sept. 6, 1997.
27. Kalwinsky DK, Raimondi SC, Schell MJ, et al. Prognostic importance of cytogenetic subgroups in de novo pediatric acute nonlymphocytic leukemia. J Clin Oncol 1990;8:75-83.
28. Gerlach H, Janisch W, Schreiber D. Intracranial and spinal tumours in newborns and infants. In: Voth D, Gutjahr P, Langmaid C, (editors). Tumour of the central nervous system in infancy and childhood. Springer-Verlag, Berlin 1982;53-7.
29. Duffner PK, Horowitz M, Krischer J, et al. Postoperative chemotherapy and delayed radiotherapy in children less than 3 years of age with malignant brain tumors. N Engl J Med 1993;328:1725-31.
30. Finlay JL, Goldman S, Wong MC, et al. Pilot study of high-dosethiotepa and etoposide with autologous bone marrow rescue in children and young adults with recurrent CNS tumors. The Children Cancer Group. J Chin Oncol 1996;14:2495-503.
31. Egeler M, de Kraker J, Voute PA. Cytosine-arabinoside, vincristine, and prednisolone in the treatment of children with disseminated Langerhans Cell Histiocytosis with organ dysfunction: experience at a single institution. Med Pediatr Oncol 1993;21:265-70.
32. Kimber C, Michalski A, Spitz L, Pierro A. Primitive neuroectodermal tumours: anatomic location, extent of surgery and outcome. J Pediatr Sung 1998;33:39-41.
33. Chan GCF, Nicholls JM, Lee CW and Lau YL. Malignant peripheral neuroectodermal tumour in an infant with neurofibromatosis type 1. Med Pediatr Oncol 1996;26:215-9.
34. Raney RB, Gehan EA, Hays DM, Tefft M, Newton WA, Heberlen V, Maurer HM. Primary chemotherapy with or without radiation therapy and/or surgery for children with localized sarcoma of the bladder, prostate, vagina, uterus, and cervix. A comparison of the results in Intergroup Rhabdomyosacrcoma Studies I and II. Cancer 1990;66:2072-81.
This web site is sponsored by Johnson & Johnson (HK) Ltd.