Intracystic bleomycin therapy for craniopharyngioma in children

The Canadian experience




Surgical removal and radiation therapy are associated with significant risk for morbidity in the pediatric population with craniopharyngioma. Intracystic therapies have been utilized in some centers to potentially decrease morbidity associated with cystic craniopharyngioma. The aim of the study was to review the Canadian experience with intracystic bleomycin therapy (ICB).


All centers in the Canadian Pediatric Brain Tumor Consortium (CPBTC) were invited to participate in a retrospective review of this treatment. A questionnaire was sent to each center. The data were analyzed at British Columbia's Children's Hospital.


In all, 6 of 17 centers utilized ICB and submitted data. A total of 17 of 19 patients with the intention to treat received bleomycin. Twelve were treated at the time of diagnosis, and 5 at the time of recurrence. Five patients achieved a complete response, 6 achieved a partial response, and 5 achieved a minor response to bleomycin. One patient was stable for 2.8 years. At the time of last follow-up, 8 patients have not required further intervention. Complications included transient symptomatic peritumoral edema (2 patients), precocious puberty (1 patient) and panhypopituitarism (2 patients). The median follow-up was 4 years (range, 0.5–10.2 years). The median progression-free survival was 1.8 years (range, 0.3–6.1 years). One patient died of a massive infarct secondary to radiation-induced moyamoya syndrome.


ICB was found to be well tolerated in this group of children. ICB may be a feasible and effective therapy for certain children with craniopharyngioma. Bleomycin may delay the need for aggressive surgery or radiation therapy for several years. Prospective multiinstitutional clinical trials are required to further evaluate the feasibility, effectiveness, and dose schedules of this treatment. Cancer 2007. © 2007 American Cancer Society.

Craniopharyngioma represents 1.5% to 11.7% of pediatric brain tumors, with an annual incidence in the U.S. of 0.18 per 100,000 population per year.1, 2 Childhood presentation is usually between 5 and 14 years of age. The tumor is histologically benign, but cure is difficult to achieve. Surgical resection is the treatment of choice for craniopharyngioma. However, even after macroscopic total resection, recurrence rates vary between 5% and 50%.3, 4 After subtotal resections, up to 70% of patients develop disease progression.5 After combination surgery and radiation therapy, disease progression or recurrence occurs in up to 20% of patients at 10 years and 46% at 20 years.6–9 Craniopharyngioma has an intimate relation with the internal carotid arteries; anterior cerebral arteries; the pituitary gland; the hypothalamus; and the optic nerves, chiasm, and tracts. Conventional therapies of surgery and radiation individuallyor in combination may be associated with considerable immediate and long-term side effects, including visual loss, endocrine dysfunction, hypothalamic dysfunction, cerebral infarction, moyamoya syndrome, cognitive dysfunction, and epilepsy.

One of the characteristics of craniopharyngiomas is the presence of single or multiple cysts, which can account for 80% to 90% of the total bulk of the tumor.10 Sixty-nine percent of evaluated patients have mainly cystic tumors.11 For the patients with predominantly cystic craniopharyngiomas there is the potential for intracystic treatment, either using chemotherapy with bleomycin; immunotherapy with interferon-α;12 or radioisotope therapy13, 14 with P32, yttrium90, or I131. To our knowledge, the role of these intratumoral modalities and more specifically the role of intracystic bleomycin (ICB) have not been well defined. The experience with ICB in the treatment of craniopharyngioma in children is limited and reported anecdotally or at the institutional level, raising questions regarding bias pertaining to published experience. The aim of this study was to review the Canadian experience and identify technical issues, outcomes, and complications associated with treating children with cystic craniopharyngioma with ICB (received from Faulding (Canada) Inc., Kirkland, Québec, Canada).


All centers in the Canadian Pediatric Brain Tumor Consortium (CPBTC) were invited to participate in a retrospective review of the use of intratumoral bleomycin in patients with craniopharyngioma up to 2003, with the most recent follow-up extending to 2004. After ethics board approval at each participating center, a questionnaire was sent to each center. Informed consent to treatment was obtained from all patients and guardians before treatment. All patients with a diagnosis of craniopharyngioma were reviewed in detail for patient age at diagnosis, cyst number and size, location of calcification if present, location of craniopharyngioma, previous therapy, clinical status, complications of catheter insertion, dosage of bleomycin, complications of bleomycin, response to bleomycin, subsequent therapy required for cranio-pharyngioma, and overall survival. A course of bleomycin was considered to be repeated doses of ICB, the frequency of which varied from daily to weekly. The dosage was noted, the total mg of bleomycin per kg per week was calculated, and the estimated concentration of bleomycin in the cyst at each instillation was also determined. Complications of bleomycin were graded according to the Common Terminology Criteria for Adverse Events, version 3.0 (CTCAE).15 The baseline scan was considered as the imaging study performed before bleomycin instillation and postinsertion of catheter. All patients had follow-up scans performed at the end of a course of bleomycin and at regular intervals subsequently. The measurements were performed by the investigator at each center. Response and progression criteria of the cystic and solid components were classified as follows: minor response (MR), >25% decrease in size; partial response (PR), >50% decrease; complete response (CR), >90% decrease; and disease progression, >25% increase in size or clinical deterioration due to increase in tumor size requiring intervention. The data were analyzed at British Columbia Children's Hospital.


Patient Population

Of the 17 centers in the Canadian Pediatric Consortium, 1 center had utilized ICB since 1995. In another 5 centers, ICB was utilized from 1999, 2000, and 2001. At 6 centers, there were a total of 42 children diagnosed with craniopharyngioma in the aforementioned time period. Of the 19 children with intention to treat, 17 actually received bleomycin therapy. Two patients (1 who was newly diagnosed and 1 who was studied after the failure of first-line therapy) had an intracystic catheter inserted, but on injection of contrast into the catheter there was leakage of contrast outside the cyst noted on computed tomography (CT) and therefore no bleomycin was instilled. One of these children was found to have communication between the ventricular system and the cyst. Twelve patients were treated with bleomycin at the time of diagnosis and 5 were treated after failing first-line therapy. ICB was not considered at the time of diagnosis in 29 of 42 of these patients because they were older, the risk of surgical resection was considered low, or the tumor was mainly solid.

The median age at the time of bleomycin therapy was 6 years (range, 1–14 years); there were 10 female and 7 male patients. Eleven of the 17 patients had tumors with a single cyst. There was a wide variation in the volume of the cyst. The median estimated volume of the largest cyst for each patient was 58 mL (range, 2.5–296 mL) (Table 1). A total of 9 of the 17 patients had tumors with a calcified cyst wall. The solid component was small, measuring ≤1 cm in greatest dimension in 14 patients, and the maximum dimension of the solid component was 2 cm. This reflects the selection bias of suitable patients by the treating centers, based on previous publications reporting the benefit of ICB in tumors that were mainly cystic.16 A total of 16 of 17 patients had tumors with a suprasellar retrochiasmatic component.

Table 1. Summary of Bleomycin Therapy
PrefixAge at diagnosis, y/GenderCyst, mLDose, mgConcentration, mg/mL/doseTotal dose per course, mgResponse of cyst (PFS, y)
  1. PFS indicates progression-free survival; D, diagnosis; M, male; mwf, Monday, Wednesday, Friday; CR, complete response; F, female; R, recurrence; PR, partial response; MR, minor response.

1 D3/MMultiple, 1002 mg, mwf0.0230CR (3.8)
6 mg/wk
2 D6/FSingle, 125 mg, mwf0.4255CR (6.1+)
15 mg/wk
3 D7/FMultiple3 mg, mwf 15CR (0.5)
9 mg/wk
4 R5/FSingle, 2.55 mg, q3d220CR (2+)
15 mg/wk
5 R8/FMultiple, 323 mg, mwf0.0945CR (0.6+)
Calcified9 mg/wk
6 D14/FSingle, 585 mg, mwf0.0945PR (3.5)
Calcified15 mg/wk
7 D3/FSingle, 1143 mg, mwf0.0336PR (0.8)
Calcified9 mg/wk
8 D2/MMultiple, 1723 mg, mwf0.0275PR (0.5+)
Calcified9 mg/wk
9 R5/MSingle4 mg wkly 8PR (3.6)
10 R4/MMultiple, 41 mg first dose then 5 mg q2d 20 mg/wk1.2561PR (4+)
11 D7/FSingle, 162.5 mg, mwf0.1630PR (0.5)
Calcified7.5 mg/wk
12 D9/MSingle, 1105 mg, mwf0.0545MR (0.3)
15 mg/wk
13 D7/FSingle, 292 mg, mwf0.0724MR (0.5)
6 mg/wk
14 D12/MSingle, 132 mg, mwf0.1536MR (0.7)
Calcified6 mg/wk
15 D2/FSingle, 1022 mg, mwf0.0230MR (0.3)
Calcified6 mg/wk
16 R5/MMultiple, 2963 mg, mwf,0.0175MR (1.8+)
Calcified9 mg/wk
17 D1/FSingle, 43 mg, mwf0.7548Stable (2.8)
Calcified9 mg/wk

Catheter Insertion

In 18 of the 19 children with intention-to-treat with bleomycin, a catheter was inserted to drain the cyst and allow bleomycin treatment, without any attempt at resecting the tumor. The approach used to insert the catheter was subfrontal in 8 patients; transcortical in 4 patients; transcortical transventricular in 3 patients; and transphenoidal, bicoronal endoscopic, and pterional in the remaining 3 patients, respectively. One patient underwent a subtotal resection at the same time that the catheter was inserted. Complications of the catheter insertion occurred in 7 of 19 patients and included acute epidural and intraventricular hemorrhage in a patient with a previously undiagnosed coagulation defect, intracystic blood (2 patients), fluid collection outside the cyst (1 patient), and contrast leakage on postoperative CT scan after instillation of contrast into the cyst (3 patients). All the children were tested for leakage outside of the cyst at least 1 week after catheter insertion, before bleomycin administration; this CT scan also provided the baseline for determining cyst size before bleomycin therapy. One child with leakage had a follow-up CT scan with contrast 1 week later that showed resolution of leakage; the patient then underwent bleomycin therapy. Five children required revision of their catheter or reservoir up to 21 months after the initial insertion.

Bleomycin Therapy

Bleomycin was administered intracystically via a subgaleal reservoir, as previously described.17 The dosage of bleomycin, clinical changes after bleomycin, and radiologic responses to bleomycin are summarized in Table 1. Treatment protocols varied between centers, with most following a Monday-Wednesday-Friday regimen. None of the patients received daily ICB. Bleomycin was administered as a course lasting an average length of 4 weeks (range, 2–15 weeks). The median dose per course was 36 mg (range, 15–75 mg). Five patients received more than 1 course of bleomycin. The median dose per instillation was 3 mg (range, 2–5 mg), and the median dose per week was 9 mg (range, 4–20 mg). The median dose of a single injection in terms of concentration within the cyst was 0.09 mg/mL/dose (range, 0.01–2 mg/mL/dose).

A CR was achieved in 5 patients, as exemplified by 1 patient whose CT scans are shown in Figure 1. Only 1 of these patients received a dose lower than the median of the whole group. Six patients achieved a PR, 5 achieved a MR, and 1 patient remained with stable disease (SD) for 2.8 years. In this latter group of patients (PR, MR, and SD), 50% received a dose of bleomycin that was below the median in terms of mg/mL/dose and of these patients who received a lower concentration of bleomycin, 66% developed disease progression within a year.

Figure 1.

(A) Axial computed tomography (CT) scan of the brain in a 3-year-old shows multiple cysts, the largest of which measured 5.8 cm in greatest dimension. (B) Axial CT scan of the brain showing a complete response after treatment with bleomycin, with a residual tumor measuring 1 cm. The solid component grew by 1 cm 43 months later, and a small cyst grew at 107 months; therefore, the patient underwent a subtotal resection followed by stereotactic radiation therapy.

Among the 17 patients treated with ICB there were 4 adverse events believed to be most likely attributable to ICB. Transient symptomatic peritumoral edema occurred in 2 patients. One presented with a decreased level of consciousness and was treated with supportive care and steroids in the intensive care unit; the edema resolved but she developed long-term panhypopituitarism. The second patient presented with multiple cranial nerve deficits and hemiparesis. She was noted to also have cyst reaccumulation in the cerebellopontine angle with brainstem compression that required revision of the catheter/reservoir system and supportive care. She had residual third nerve pareses, hemiparesis, and decreased attention long-term. One child developed precocious puberty in spite of cyst shrinkage 5 months later and 1 was noted to have panhypopituitarism at 2 years in spite of persistent resolution of the cyst. None of our patients developed secondary visual dysfunction directly due to the bleomycin and 4 experienced visual improvement after the bleomycin therapy.

Two patients developed moyamoya syndrome several years later, 1 due to tumor infiltration into the anterior cerebral arteries (5 years postradiation) and the other secondary to radiation therapy given 4.5 years previously, when the child was 8 years of age. One patient developed hyperpigmentation over bony prominences, which was considered a systemic side effect of bleomycin. The 1 patient who had long-term neurotoxicity potentially secondary at least in part to bleomycin had received the highest dose of bleomycin in terms of mg/week, mg/kg/week, and the second highest estimated concentration of bleomycin in the cyst.

In 3 patients who underwent surgical resection for progression after ICB, the pathology of the tumor was reviewed with a special focus on unusual features related to bleomycin. There were only the usual features of normal thin-walled vessels in the gliotic brain adjacent to the craniopharyngioma. The pathology was reviewed in 1 patient after both bleomycin and radiation and there were the expected changes after radiation with fibrous tissue replacement of blood vessel walls and Rosenthal fibers in adjacent gliotic brain.

Subsequent Therapy

Eight patients required no further intervention after a single course of bleomycin. Two patients underwent a second course of bleomycin only at the time of cyst recurrence. Thus, 10 patients have been treated with bleomycin alone and have been followed for a median of 2+ years (range, 0.5–6.1+ years). A sustained benefit of >1 year was observed in 11 of 17 patients for a median of 3.6 years (range, 1.6– 9.3 years) before another modality of therapy was required. One patient progressed within 3 months and underwent a subtotal resection. Two patients developed disease progression and received involved field radiation therapy. Five patients received subsequent multimodality therapy including bleomycin (3 patients), resection (5 patients), and radiation therapy (5 patients). Radiation therapy was delayed by a median of 3.6 years (range, 0.2–9.3 years).


The median progression-free survival for patients treated at the time of diagnosis with ICB was 0.7 years (range, 0.3–6.1 years) with a median overall survival of 5 years (range, 0.5–10.2 years). A single patient died; this death was secondary to a massive middle cerebral artery territory infarct secondary to moyamoya syndrome 4–5 years after radiation therapy. The median progression-free survival for patients treated at the time of recurrence with ICB was 2 years (range, 0.6–4 years), with a median overall survival of 2+ years (range, 0.6–5.5+ years); none of these patients had died at the time of last follow-up, but the follow-up period was relatively short.


In spite of the benign histology of craniopharyngioma, it can behave aggressively.18 The 10-year overall survival is 64% to 92%, which is consistent with the benign histology, regardless of whether the tumor is treated in a surgically aggressive manner or with radiotherapy or a multimodality approach.19–21 However, the acute and long-term morbidity of childhood craniopharyngioma remains substantial, with nearly universal multiple endocrinopathies, visual loss reported in up to 75% of patients, severe postsurgical obesity, a high prevalence of neurocognitive sequelae, and up to 20% of patients reported to develop neurologic motor deficits or a seizure disorder.20–24 The risks of radiation to the developing brain are much higher, thus creating the need for avoiding this modality in younger children or at least delaying its use. Therefore, a permanent obliteration of the cyst or at least a transient shrinkage could be a valuable contribution to therapy in the child with cystic craniopharyngioma. ICB has been utilized in selected cases as part of a multimodality approach, particularly in the younger child who is at greatest risk of complications from aggressive surgery and radiation therapy, and also in recurring cystic craniopharyngioma.

To our knowledge, the literature to date regarding the use of ICB for craniopharyngioma is limited to case reports or individual center retrospective series and 1 small prospective study of 5 patients. The largest series published to date included 26 patients (20 children and 6 adults) from a single center. None of these reports have provided a current review of the literature, a multiinstitutional perspective, or details of the concentration of dose within the cyst, or correlated dose to response rate; therefore, we believe the current study adds valuable information.

On review of the literature of 70 adult and pediatric patients with craniopharyngioma who were treated with ICB, it is reported to be effective in inducing at least >25% cystic shrinkage in up to 90% of patients, and a CR in 50% of patients. Approximately 50% of patients were successfully treated with bleomycin alone at a mean follow-up of 5 years (range, 0.2–16 years).12, 16, 17, 25–33

The results in our pediatric population at a median follow-up of 5 years (range, 0.5–10.2 years) are in keeping with the previously published experience: 94% had at least 25% shrinkage and 29% achieved a CR. In all, 59% were successfully treated with bleomycin alone; 2 patients underwent a repeat course of bleomycin and the median progression-free survival was 2 years (range, 0.5–6.1+ years). However, the duration of benefit was <1 year in 41% of patients, leading to more definitive therapy in 30% within 1 year. A sustained benefit of >1 year was noted in 11 of 17 patients for a median of 3.6 years (range, 1.6–9.3 years) before another modality of therapy was required, thus potentially decreasing the risks of morbidity of radiation and/or surgery. Only 1 of 9 patients with a calcified cyst had a CR; this may be an adverse factor for cyst collapse after bleomycin therapy. In addition, our comprehensive data collection suggests that at least one-third of pediatric craniopharyngioma patients present with cystic tumors that are amenable to this treatment approach.

Based on our experience and the review of the literature, it appears that dosing is complicated and does not neatly correlate with response. Our findings suggest a lower chance of significant cyst shrinkage and shorter duration of response if the patient received <9 mg/wk and 0.09 mg/mL/dose. However, we have observed that this tumor behaves in a heterogeneous manner: some large cysts will disappear with a small dose of bleomycin and small cysts may only have an MR with a higher dose. Two of 5 patients who underwent a second course of bleomycin have not required any further intervention at the time of last follow-up, suggesting that this may be a reasonable tactic in those who are young, in whom surgical and radiation risks of morbidity are high. Alternate-day dosing appears safe; our data suggest concentration and weekly dose may be important, but no one really knows the optimal individual dose. Rarely reported complications include blindness in 1 patient related to incorrect dilution31; sensorineural hearing loss,26 likely a direct toxic effect secondary to reservoir communication with cerebrospinal fluid; single reports of long-term hypothalamic dysfunction,27 focal ischemia,26 and seizure with hemiparesis28; and death possibly related to a high individual and cumulative dose, and daily dosage scheme over a short period (daily dose over 8 days, for a total of 56 mg).32 In our experience 12% of children developed transient grade 3–4 neurotoxicity, 1 developed residual panhypopituitarism, and another patient developed a residual hemiparesis and third nerve pareses. One child developed precocious puberty and a fourth child developed panhypopituitarism possibly secondary to the bleomycin treatment.

The effects of ICB on future surgical resection, if required, are unclear. In the 3 cases reviewed by our neuropathologist, there were no unusual features of the tumor or its vasculature. An autopsy was performed on 1 of the patients reported by Takahashi et al.16 This patient had a huge mixed tumor and Rosenthal fibers and microcapillary congestion was observed in the normal brain adjacent to the tumor, whereas no prominent necrotic foci were found.

To our knowledge, there are 2 reports describing the use of ICB for craniopharyngioma followed by radiation. Jiang et al28 describe 9 patients receiving 5–15 mg of ICB daily over 8 days followed by intracystic P32 at a dose of 200 grays. Three adults had severe sequelae: fatigue; hyponatremia, adephagia obesity, thalamic infarction, and death. Nicolata et al.33 reported the treatment of 8 patients with bleomycin followed by gamma knife radiosurgery. All patients had at least a 50% reduction in tumor size; 1 patient died 44 months later secondary to diffuse ventriculitis. The daily dosage of bleomycin with adjuvant radiation may be associated with greater toxicity. Our patient who developed radiation-induced moyamoya syndrome received the radiation within 10 months of the bleomycin therapy. It is unclear whether the bleomycin was a contributing factor because the vasculopathy became symptomatic 4.5 years after radiation. Moyamoya syndrome is a well-recognized complication of radiation to the suprasellar region; the vast majority of patients are children.34 We observed a benefit in our population who received bleomycin in terms of radiation deferral.

The limitations of this study are inherent due to the retrospective nature. They include selection bias based on each institution's philosophy and experience; the mix of newly diagnosed and recurrent patients at the time of ICB therapy; variability in the dosage of bleomycin and the frequency of each instillation; and the relatively short follow-up in some patients.


ICB was found to be well tolerated in the majority of children in this retrospective review. The untoward effects were minor in most instances, and in those patients who developed more severe consequences the long-term repercussions appear to be less severe than those of aggressive surgery or radiation therapy in the young child. A minority of children with newly diagnosed craniopharyngioma may be suitable for this therapy. ICB might be considered in selected cases: mainly cystic lesions, in children age <10 years at the time of diagnosis, and for lesions that are of high risk of serious surgical morbidity (eg, retrochiasmatic hypothalamic lesions). This treatment can also be considered for cystic recurrence after standard therapy. In our series, a dose >15 mg/wk or 0.9 mg/kg/wk or 1 mg/mL/dose was associated with greater toxicity. Delivering therapy on alternate days may maximize efficacy and minimize serious complications. Adjuvant radiotherapy may increase the risks of serious complications. It is important to ensure no leakage before bleomycin administration. Bleomycin may delay the need for aggressive surgery or radiation therapy for several years. Bleomycin appears to be a useful alternative in the treatment of cystic craniopharyngioma occurring in children, but prospective clinical trials are required to evaluate this further. The results of the current study provide the basis for a future prospective multiinstitutional study of childhood cystic craniopharyngioma. In view of the side effect profile, other drugs should also be considered in the treatment of this disease.


We thank Victor Espinosa of the Statistics Department at British Columbia's Children's Hospital; and Vesna Popovska, research support, Neurology Division, British Columbia's Children's Hospital.