Introduction

Since the introduction of the routine use of central nervous system (CNS) prophylaxis for children with acute lymphoblastic leukemia (ALL), the cure rates have been markedly improved [20]. With the prolongation of the survival time, undesirable results of these protocols, such as secondary malignancies, have been identified. In this report, we describe a rare case of a child who developed an anaplastic oligoastrocytoma of the left parietal lobe 9 years after the successful treatment of ALL.

Case report

A 14-year-old male patient admitted to our neurosurgery clinic with a history of a single episode of generalized tonic–clonic seizure, which occurred 3 weeks ago. Neurological examination revealed nothing abnormal. He had a history of ALL treated 9 years ago with induction chemotherapy, intrathecal (IT) methotrexate (Mtx) and prophylactic whole brain irradiation (1,800 cGy in 10 fractions over 2 weeks). The magnetic resonance imaging (MRI) that was performed 2 weeks before his admission revealed two well-demarcated lesions localized in the left posterior parietal and right frontal regions (Fig. 1a–c). The patient was hospitalized for further investigation. He was consultated with pediatric hematology and oncology. The bone marrow and peripheral blood examinations showed no evidence of ALL recurrence. Cerebrospinal fluid studies (biochemistry, cytology, oligoclonal IgG) and blood biochemistry were in normal ranges. Acute disseminated encephalomyelitis or malignancy was suspected in the differential diagnosis. The second cranial MRI scan, which was performed approximately 4 weeks after the first, revealed a rapid progression of both lesions (Fig. 2a–c). He then underwent a left frontoparietal craniotomy, and the lesion located at the left parietal lobe was totally removed (Fig. 3). The postoperative period was uneventful, and the patient was discharged on the sixth postoperative day. The pathological diagnosis was anaplastic oligoastrocytoma grade III (Fig. 4). He was referred to pediatric hematology–oncology department for further treatment.

Fig. 1
figure 1

a T2-weighted axial MRI scan demonstrates two well-demarcated lesions (arrows) localized in the left posterior parietal and right frontal regions. b T1-weighted contrast-enhanced coronal and c T2-weighted sagittal MRI images

Full size image
Fig. 2
figure 2

T1-weighted contrast-enhanced MRI scans a axial, b coronal, and c sagittal sections clearly reveals rapid progression

Full size image
Fig. 3
figure 3

Postoperative cranial CT shows total resection of the tumor

Full size image
Fig. 4
figure 4

Photomicrograph (original magnification ×200 hematoxylin–eosin). Highly cellular tumor with both oligodendroglial (∼35%) and astrocytic (∼65%) components. Typical neoplastic oligodendrocyte with a round nucleus and a perinuclear halo (thin arrow); a neoplastic astrocyte with pleomorphic, hyperchromatic nuclei (thick arrow); pleomorphism (repeating short arrows). No necrosis or microvascular proliferation was identified. These features are consistent with an anaplastic oligoastrocytoma (WHO grade III)

Full size image

Discussion

Central nervous system (CNS) prophylaxis is an essential component of the management of ALL in children. The treatment modalities to prevent CNS leukemia include cranial irradiation, IT chemotherapy, and high-dose systemic chemotherapy. With the use of these protocols, the cumulative incidence of CNS leukemia has been reduced from 30–50% to less than 10% [4, 20]. Among these treatment modalities, cranial irradiation has been shown to have serious complications both in short term and in long term. These side effects, such as white matter destruction, vascular damage leading to calcifications, and enlargement of ventricles and/or sulci secondary to cortical atrophy, have all been identified in radiologic examinations [22]. Clinically, postradiation damage represents a broad spectrum of signs and symptoms such as intellectual dysfunction, learning disabilities, growth retardation, neurological sequelae, seizures, and secondary neoplasia [4, 22]. Due to these effects, recent CNS prophylaxis regimens have shifted from cranial irradiation to IT plus high-dose systemic chemotherapy.

Radiation-induced neoplasia is a well-recognized condition, in which the meningiomas and sarcomas predominate among other tumor types [12, 23]. In 1950s, a retrospective survey of 11,000 children irradiated for tinea capitis revealed a significantly higher risk of both benign and malignant head and neck tumors [12]. In the majority of these cases, the interval between radiation and tumor diagnosis was more than 9 years. Possible radiation-induced gliomas following ALL treatment have also been reported in the literature, and this concomitant presentation has been investigated [1, 3, 6–8, 10, 13, 15, 18, 24–26].

A relationship between young age at ALL diagnosis and increased occurrence of high-grade gliomas has been proposed in large clinical studies [9, 24]. Walter et al. [24], in their series of 1,612 patients consecutively treated for childhood ALL with a median follow-up of 15.9 years, evaluated the incidence and potential risk factors for secondary malignant neoplasms of the brain. The cumulative incidence of brain tumors among these patients was reported to be 1.39% at 20 years. Of the 21 patients with 22 brain tumors, 10 were high-grade gliomas, 1 was a low-grade glioma, and 11 were meningiomas. Gliomas occurred at a median latency of 9.1 years in contrast to longer latency of meningiomas (19 years). The risk factors for developing any secondary brain tumor included presence of CNS leukemia at diagnosis and increased cranial irradiation dose. Young age at ALL diagnosis (<6 years) was found to have an association with an increased risk of developing a high-grade glioma. Nine out of 10 patients with high-grade gliomas in this study were younger than 6 years old at the time of ALL diagnosis [24]. In a multicenter study enrolling 5,006 children treated for ALL, 13 secondary CNS neoplasms were detected [9]. Likewise, tumors of the CNS were most commonly found to occur among children below the age of 7 at the time of ALL diagnosis. The cumulative probability that a CNS tumor would develop in these patients was 1.5% after 15 years, significantly higher compared with the risk of 0.1% in patients 7 years of age or older [9]. Our patient was 5 years old at the time of ALL diagnosis, and this may have played an important role in the oncogenic process. Previously reported, fully documented cases of secondary glioma following ALL treatment in patients who were less than 6 years old at the time of ALL diagnosis are summarized in Table 1.

Table 1 Previously reported, fully documented cases of CNS glioma following ALL treatment, with focus on age at first ALL diagnosis (<6 years old)
Full size table

We believe that relative immaturity of the brain at the time of radiation therapy in ALL patients can be a predisposing factor for secondary CNS neoplasia. Traditionally, CNS has been described as relatively resistant to radiation-induced changes. However, in young children, the turnover of glial cells, especially in the early postnatal period, is more intense, making them potential targets for oncogenesis. This was stressed by Palma et al. [13] who speculated that the child’s brain is susceptible to the mutagenic effects of ionizing radiation particularly during myelinogenesis.

Several criteria must be fulfilled in order to accept radiation as a possible etiological factor for neoplasia [2, 19]. These are as follows: (1) the tumor must occur in the irradiated area, (2) the tumor must not be present prior to radiation, (3) a sufficient latency period must elapse between irradiation and the occurrence of the tumor, (4) the radiation-induced tumor should differ histopathologically from the original lesion, and this must be proven. Even though our case fulfils all of the above-mentioned criteria, the definite factors participating in the development of secondary neoplasia remain to be speculative. The evaluation of the second cranial MRI, which was performed approximately 4 weeks after the first investigation, demonstrated the aggressiveness of the tumor, and the factor triggering this rapid progression also remains to be unknown.

Although radiotherapy is clearly a strong risk factor, the occurrence of glioma in ALL patients who have not received such treatment [16, 21] prompted the evaluation of other factors. Genetic susceptibility and linkage between brain tumors and hematologic malignancies has been proposed [9–11]. Meadows et al. [11], in their clinical study of 102 children, each harboring more than one malignant neoplasm, reported that there was a notable association of glioma with leukemia or lyphoma. The authors suggested that this finding may be associated with “a new cancer syndrome,” and certain patients with ALL might be genetically susceptible to the development of gliomas [11]. Although lacking definitive proof, increased incidence of CNS tumors among the first-degree relatives of patients suffering from leukemia supports this hypothesis [5, 8]. Systemic chemotherapy agents, such as thiopurines and antifolates, have also been suggested to participate in the oncogenetic processes [17], and IT Mtx has been accused for increasing the risk of glioma development [6, 17]. Mtx, when administered in combination with 200 cGy or more cranial irradiation, has been shown to cause multiple necrotic areas and leukoencephalopathy [14]. However, there is no single evidence both in man and experimental animals to indicate that Mtx alone is carcinogenic or that it enhances the carcinogenic effect of radiation. Moreover, the use of intrathecal chemotherapy was not found to increase the risk of developing secondary brain tumor in a clinical study [24]. Depending on the pertaining literature, we think that the oncogenic process of secondary CNS malignancy in ALL patients is multifactorial, and synergistic action of the previous treatment modalities plays a major role.

In conclusion, children suffering from ALL and treated by prophylactic CNS irradiation and IT Mtx can present with intracranial tumors even long after the completion of the initial treatment. Although the tumor would most likely to be a high-grade glioma in these cases, the possibility of an oligo component should not be ignored. Since early detection of the intracranial lesions may permit a safe surgical removal, regular follow-up with radiological investigations should be performed systematically, especially if a personal history of cranial radiotherapy is present.