To the Editor:
The potential carcinogenesis induced by radiation and the occurrences of neoplasms after radiation therapy are well-accepted facts.1 Any report on radiation-induced tumors must first begin with Cahan's criteria.2 Established in 1948, these criteria were thought to be essential to classify a tumor as radiation-induced. Specifically, the second tumor must occur within the original radiation field but not have been present on imaging at the time of initial irradiation; there must be a latency period between the radiation exposure and the development of the second tumor; the second tumor must be histologically unique from the original tumor; and the patient cannot have a genetic syndrome that predisposes them to cancers.
In the recently published articles, Sofela et al3 and Signorelli et al4 reported cases of malignant transformation (MT) of craniopharyngiomas. Although the case described by Sofela et al fulfilled Cahan's criteria, their literature review showed a poor correlation between radiation therapy and MT of craniopharyngiomas. Likewise, Signorelli et al added their case in which radiotherapy was not performed to the Sofela et al series and compared it with other neurosurgical series of craniopharyngiomas. They also suggest with their statistical analysis that radiotherapy has no role in MT of craniopharyngiomas.
We read these recent reports with great interest because we have also experienced a case of an aggressive MT of a craniopharyngioma. The 40-year-old man was first seen at our institution in 2012 with a growing tumor and history of multiple surgeries and radiation treatments. The patient had a partial resection of a craniopharyngioma via a microscopic sublabial approach followed by radiation therapy at another institution in 1992 when he was 20 years old. Twelve years later, in 2005, he was diagnosed with adenoid cystic carcinoma of the left parotid gland and underwent a radical resection followed by intensity-modulated radiation therapy and chemotherapy. In 2011, 19 years after the first surgery, the patient had a recurrence of the craniopharyngioma and underwent surgery a microscopic transnasal transsphenoidal approach with a partial resection of the lesion. Pathology revealed a malignant craniopharyngioma with squamous cell carcinoma deterioration. He received chemotherapy and proton radiation therapy for the residual lesion.
In 2012, the patient presented with a growing ventral skull base lesion involving the posterior, middle, and part of the anterior cranial fossa (Figure, A and B). He was neurologically intact with the exception of left-sided vision loss and facial nerve palsy from his previous surgeries. We performed an expanded endoscopic endonasal approach with an aggressive subtotal resection, leaving macroscopic residual tumor on the left cavernous sinus (Figure, C and D). Pathology confirmed a squamous cell carcinoma, and the patient underwent multiple chemotherapy trials that controlled the lesion growth. However, the chemotherapy was discontinued after 1 year due to tumor progression. The tumor progression involved all of the ventral skull base and nasopharynx (Figure, E and F). We offered the patient further surgical removal of the tumor as a palliative management. Surgery using an expanded endoscopic endonasal approach was performed to remove the midline ventral skull base component of the tumor. Five days later, a left pterional craniotomy was performed to remove the parasagittal component of the lesion, and a near-total resection was achieved (Figure, G and H) with no new neurological deficit. Unfortunately, 3 months after the last surgery, the tumor recurred, and the patient opted for hospice care.
After experiencing this aggressive MT of craniopharyngioma and reviewing the literature, we would like to point out some aspects of this extremely rare condition.
Definite diagnostic criteria for malignant craniopharyngioma have not been established. In the previous literature, malignant craniopharyngioma appears to present the following features: cellularity and an increased nuclear cytoplastic ratio; nuclear pleomorphism and hyperchromatic nuclei; increased mitotic activity or higher proliferating cell nuclear antigen expression; coagulative necrosis; and other histologic features such as a solid growing pattern, destruction of the basement membrane, infiltrative growth, microvascular proliferation, and so on. It seemed that previously reported malignant craniopharyngiomas always included 3 or more of these features.
Despite the statistical analysis showing poor correlation between radiation therapy and MT of craniopharyngiomas, we respectfully disagree with the conclusions of Sofela et al3 and Signorelli et al4 that radiation plays a minor role, if any, in the pathogenesis of malignant craniopharyngiomas. In addition to their series, there are 2 other cases (including our report) of MT that occurred after radiation therapy.5 In total, there are 16 cases of MT of craniopharyngioma related to previous radiation and 4 cases of MT with no radiation history. These numbers are still insufficient to find a statistically significant correlation. However, the difference in the number of cases with and without previous radiotherapy cannot be ignored based on the potential carcinogenesis effect of radiation and the rarity of this malignancy. We might be facing a type II statistical error due to a small sample size, leading us to wrong conclusions.
The most robust evidence of central nervous system neoplasms induced by radiation comes from the Israeli experience with tinea capitis.6 In the post–World War II era, 10 834 children received scalp irradiation for the management of tinea capitis. The average dose to the brain was estimated at 1 to 2 Gy per treatment, with a mean dose of 1.5 Gy for the cohort. At 30 years post-exposure, a central nervous system malignancy had developed in 60 of the 10 834 children, which represents an increase in relative risk of 8.4 over the general population. Of note, the time to development of tumors depended on histology; malignant tumors developed earlier (∼9-14 years after radiotherapy), whereas benign tumors developed later (∼15-20 years after radiotherapy).6
In addition to the tinea capitis experience, numerous other studies support the concept of radiation-induced central nervous system malignancies. External beam radiotherapy for pituitary tumors has been shown to increase the risk of meningiomas, gliomas, and sarcomas.7 Likewise, external beam radiotherapy for childhood cancers is well documented to increase the risk of central nervous system malignancies, often in a dose-dependent fashion. Overall, the published literature quotes a 1% to 3% risk at 10 to 20 years of the development of a secondary neoplasm after traditional fractionated radiotherapy.7 The take-home message from these reports is that large volumes of low-dose radiation to normal structures significantly increase the risk of secondary neoplasms.7
The cytotoxic and mutagenic effects of radiation might explain this increased risk. If the radiation dose delivered is too high, the normal cells will simply die and have no opportunity to become neoplastic. Animal and clinical studies have demonstrated that there is an increasing rate of secondary neoplasm development up to a maximum dose between 3 and 10 Gy, followed by a decrease in risk as the dose increases beyond this range.8,9
Although MT of craniopharyngiomas is rare and no statistical significance related to radiation was found, it is hard to believe that this treatment modality is innocuous. Even though the recent methods to deliver radiation, such as stereotactic radiosurgery and intensity-modulated radiation therapy, decrease the dose delivered to normal tissue, the risk of the development of radiation-induced tumors is not zero.
Currently, our philosophy is to carefully select the patients (in special children) with a benign histology who will receive radiation therapy. Usually we opt to deliver radiation in cases of recurrent or growing residual tumor not amenable to surgical resection.
In conclusion, we believe that there is enough evidence in the literature to suggest the association between radiation and malignant transformation of craniopharyngiomas. The use of radiation on the management of craniopharyngiomas should be parsimonious and ideally reserved for cases of confirmed disease progression. The risks of radiotherapy side effects, including MT, must be discussed fully with patients who have histologically benign tumors.
The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
1. Kohn HI, Fry RJ. Radiation carcinogenesis. N Engl J Med. 1984;310(8):504–511.
2. Cahan WG, Woodard HQ, Higinbotham NL, Stewart FW, Coley BL. Sarcoma arising in irradiated bone; report of 11 cases. Cancer. 1948;82(1):8–34.
3. Sofela AA, Hettige S, Curran O, Bassi S. Malignant transformation in craniopharyngiomas. Neurosurgery. 2014;75(3):306–314.
4. Signorelli F, D'Alessandris QG, Maira G, Pallini R, Lauretti L. Letter: malignant craniopharyngioma and radiotherapy: the missing link. Neurosurgery. 2015;76(3):E358–E359.
5. Wang W, Chen XD, Bai HM, et al. Malignant transformation of craniopharyngioma with detailed follow-up. Neuropathology. 2015;35(1):50–55.
6. Ron E, Modan B, Boice JD Jr, et al. Tumors of the brain and nervous system after radiotherapy in childhood. N Engl J Med. 1988;319(16):1033–1039.
7. Patel TR, Chiang VL. Secondary neoplasms after stereotactic radiosurgery. World Neurosurg. 2014;81(3-4):594–599.
8. Epstein R, Hanham I, Dale R. Radiotherapy induced second cancers: are we doing enough to protect young patients? Eur J Cancer. 1997;33(4):526–530.
9. Niranjan A, Kondziolka D, Lunsford LD. Neoplastic transformation after radiosurgery or radiotherapy: risk and realities. Otolaryngol Clin North Am. 2009;42(4):717–729.