Treatment Strategies for Localized Central Nervous System Germinoma

Sep 22, 2010

Combined Chemotherapy and Radiation Versus Radiation Alone: Two Opposing Views

Amar Gajjar, MD
St. Jude Children’s Research Hospital

October 2010 Issue: Primary intracranial germ cell tumors represent 3% of all intracranial tumors in children in Europe and the United States, with a higher prevalence in Japan and Asia. The peak incidence is in the second decade of life. Germinomas account for approximately twothirds of intracranial germ cell tumors. Progression-free survival in pure germinomas is excellent, with five-year overall survival (OS) rates exceeding 90% in many retrospective and prospective series.

Historically, craniospinal irradiation (CSI) followed by a boost to the primary tumor area has been regarded as the standard treatment for intracranial germinoma. However, the role of CSI in the treatment of localized germinoma has been called into question due to its late effects and the low incidence of spinal relapses in series that have omitted spinal irradiation. Excellent responses have been achieved with whole brain radiation therapy and whole-ventricle irradiation followed by a boost to the primary tumor(s).

Systemic chemotherapy followed by involved field irradiation has been tested as a means of reducing radiation doses and/or volume while maintaining high cure rates. Current prospective protocols are seeking to maintain the high cure rates for this tumor while developing therapy with the fewest short- and long-term toxicities.

The following represents two independent points of view: the merits of combined modality therapy, presented by Eric Bouffet, MD, compared with the benefits of treatment with CSI therapy, presented by Rolf Kortmann, MD. Important areas of discussion include contemporary staging of newly diagnosed patients with central nervous system (CNS) germinoma, cure rates with different treatment modalities alone and in combination, dose and volume of irradiation, and toxicities and late effects of each treatment approach.


  1. Finlay J, da Silva NS, Lavey R, et al. Pediatr Blood Cancer. 2008;51:313-6.

Combined Chemotherapy and Limited-field Radiation Therapy

Eric Bouffet, MD, FRCP(C)
The Hospital for Sick Children Toronto, Canada

The relative contributions of radiation therapy and chemotherapy remain a topic of debate in the treatment of highly curable CNS germinomas. The efficacy of chemotherapy in patients with intracranial germ cell tumors has been recognized since the mid-1980s,1 and cyclophosphamide, cisplatin, carboplatin, and etoposide have demonstrated activity in patients with newly diagnosed and recurrent tumors. In a pilot study of pre-radiation carboplatin for patients with intracranial germinoma, seven complete and three partial responses were observed in 10 evaluable patients.2 In another study of neoadjuvant (pre-radiation therapy) chemotherapy, the response rate to cyclophosphamide was excellent, with seven complete responses out of eight patients with germinoma.3 Although all of these drugs appear to be highly effective, the combination of carboplatin and etoposide appears to be the gold standard in the management of germinoma, as chemotherapeutic agents that require hyperhydration can be associated with significant side effects in patients with diabetes insipidus.4,5

Minimizing adverse effects
The efficacy of chemotherapy observed in pilot studies has contributed to introducing pre-irradiation chemotherapy to treat patients with germinoma in order to allow a reduction in irradiation treatment volume and doses. The main objective of this combined treatment approach is to minimize potential long-term effects of large-volume, high-dose radiation—such as neurocognitive deficits, vascular complications, and endocrine deficits—particularly among patients with nonmetastatic germinoma. However, there currently is no consensus on an optimal treatment strategy, and the absence of large prospective studies with long-term follow-up precludes any definitive conclusion.

Studies confirming the combined approach
The feasibility and the efficacy of combining chemotherapy with limited-field radiation have been confirmed in several studies. Using a combination of carboplatin, etoposide, and ifosfamide as initial therapy, followed by a radiation treatment to the initial tumor bed with a 1-cm to 2-cm margin at a dose of 4,000 cGy, a study reported a 93.3% event-free survival and 100% OS at 32 months in a group of 29 patients.6 These results were confirmed in a follow-up report of 57 patients by the same investigators.7 Others have reported outcomes of newly diagnosed patients receiving lower doses of irradiation. Seventeen patients were treated with four cycles of either etoposide and cisplatin or ifosfamide/etoposide/cisplatin (ICE) followed by 2,400 cGy focal irradiation in 12 fractions.8 Three patients with dissemination received CSI. At a median follow-up of 24 months, 16 of 17 patients were alive without recurrence.

Following these early studies, the Japanese Germ Cell Tumor (GCT) Study Group conducted a cooperative study (1995-2003) using a combination of carboplatin (450 mg/m2 on day 1) and etoposide (450 mg/m2 on days 1-3) for three courses followed by focal radiation at a dose of 24 Gy for patients with nonmetastatic germinomas. One hundred and twenty-three patients with germinoma were enrolled, and the five-year OS was 98%. However, three cooperative groups (Japanese, French, and the Societé Internationale Oncologie Pédiatrique [SIOP]) achieved similar conclusions when they reviewed the pattern of relapse among patients treated with focal irradiation and chemotherapy9-11 and reported a 10% to 15% relapse rate using this approach. Importantly, a significant number of relapses occurred in the vicinity of the primary tumor, particularly within the ventricular region. All groups concluded that focal radiation was associated with an increased risk of ventricular failure and that whole ventricular field irradiation should be the “standard” approach for patients with nonmetastatic germinoma.

Ongoing studies in the SIOP group and in Japan are aimed at confirming the safety and relevance of this approach. This combined strategy only concerns patients with nonmetastatic germinoma, and the benefit of pre-radiation chemotherapy for patients with evidence of dissemination who receive craniospinal radiation is still unproven. However, this combined approach seems to be applicable to patients with bifocal tumors, and excellent outcomes have been reported with chemotherapy followed by limited-field radiation in this subgroup, which accounts for 15% to 25% of all germinomas.12,13

Unanswered questions and future studies
Overall, the introduction of chemotherapy in the management of intracranial germinoma has contributed to a significant reduction in both dose and volume of radiation. Pilot studies have shown that a dose of 24 Gy and a limited field are sufficient when pre-radiation chemotherapy is used. There is currently no evidence that more intensive chemotherapy will allow further volume or dose reduction of radiation. Although current outcomes are excellent, there are still unanswered questions. Future studies should try to better identify the risk of ventricular spread of pineal and/or suprasellar germinoma in order to tailor radiation fields and volumes accordingly.


  1. Allen JC, Bosl G, Walker R. J Neurooncol. 1985;3:147-52.
  2. Allen JC, DaRosso RC, Donahue B, et al. Cancer. 1994;74:940-4.
  3. Allen JC, Kim JH, Packer RJ. J Neurosurg. 1987;67:65-70.
  4. Afzal S, Wherrett D, Bartels U, et al. J Neurooncol. 2010;97:393-9.
  5. Kellie SJ, Boyce H, Dunkel IJ, et al. Pediatr Blood Cancer. 2004;43:126-33.
  6. Baranzelli MC, Patte C, Bouffet E, et al. Cancer. 1997;80:1792-7.
  7. Bouffet E, Baranzelli MC, Patte C, et al. Br J Cancer. 1999;79:1199-204.
  8. Sawamura Y, Shirato H, Ikeda J, et al. J Neurosurg. 1998;88:66-72.
  9. Alapetite C, Ricardi U, Saran F, et al. Med Pediatr Oncol. 2002;39:248.
  10. Calaminus G, Nicholson JC, Alapetite C, et al. Med Pediatr Oncol. 2002;39:227.
  11. Matsutani M, ed. Treatment of intracranial germ cell tumors: the second phase II study of Japanese GCT group . Abstracts from the Thirteenth International Symposium of Pediatric Neuro-oncology, Chicago: Neuro-oncology. 2008;10:420.
  12. Lafay-Cousin L, Millar BA, Mabbott D, et al. Int J Radiat Oncol Biol Phys. 2006;65:486-92.
  13. Lee L, Saran F, Hargrave D, et al. Childs Nerv Syst. 2006;22:1513-8. 18 I


Craniospinal Irradiation Alone in Pure Germinoma

Rolf D. Kortmann, MD
University of Leipzig
Leipzig, Germany

Radiation therapy in pure germinoma is indispensable in providing a cure. Chemotherapy alone cannot replace radiation therapy as sole treatment and can only be performed in a combined setting.1-3 Additionally, a reliable and complete staging—including histologic verification, assessment of tumor markers, cerebrospinal fluid cytology, and brain and spinal magnetic resonance imaging to exclude non-seminomatous germ cell tumors and metastatic spread—is mandatory. The current recommended dose to treat subclinical disease is 20 Gy to 24 Gy and 40 Gy to macroscopic tumor, independent of the addition of chemotherapy.4

Radiation therapy alone in localized disease
The literature is replete with retrospective studies reporting on radiation therapy alone in pure intracranial germinoma using different treatment volumes. When reviewing the data, it becomes clear that any reduction in treatment volumes is notoriously associated with increased relapse rates, especially if staging is incomplete. Presently, the most important issue in radiation therapy is whether a reduction of treatment volumes can be afforded without the supplemental use of chemotherapy.

Craniospinal irradiation
CSI alone is a consequent treatment and has long been considered the gold standard, providing cure in more than 95% of patients. Rogers and colleagues reviewed published data on 343 patients who received craniospinal radiation therapy as sole primary treatment.5 Local control was achieved in all but one patient, and four (1.2%) had isolated spinal relapses. Overall there were 13 relapses (3.8%), and more than half of the reported relapses occurred outside the craniospinal axis. However, adverse late effects have been suggested, particularly in younger patients, raising the question of whether reduced treatment volumes could be applied without jeopardizing the high cure rates and whether additional chemotherapy could be avoided, as the combination of both treatment modalities may cause increased toxicities to the CNS and lead to an increased rate of secondary malignancies as suggested in a large-scale SEER analysis.6

Whole brain/whole ventricular system irradiation
In modern studies that have applied various volumes to intracranial germinomas, authors tend to conclude that if staging is complete and localized disease is confirmed, spinal radiation therapy may be safely omitted. Studies with no prophylactic irradiation of the spine (in which complete staging was not obtained for all patients) reported rates of relapse to the spine have been low: 0% to 10%.7-14 In one study, the incidence of spinal relapses was 4% (two of 56) for patients treated with spinal irradiation and 3% (two of 70) for those without spinal irradiation.14 In a retrospective study, local field and/or whole brain irradiation was performed in 114 patients and CSI in 66 patients. Eight-year overall and event-free survival rates were 91% and 89%, respectively. The eight-year recurrence rates at the primary site, intracranial space, and the spinal space were 1%, 6%, and 6%, respectively, regardless of target volume.15 In 278 patients receiving either whole brain or whole ventricular irradiation, local control was achieved in 97.5%.5 The frequency of isolated spinal relapses (eight patients) did not differ significantly from that for CSI (2.9% vs. 1.2%).

When whole ventricular irradiation has been used, extra-ventricular intracranial relapses have rarely been reported, suggesting that whole brain radiation therapy would not be necessary. One study showed that no isolated spinal cord relapses occurred in 41 patients with localized germinoma. Six patients received CSI and 35 did not. Twenty-one patients with localized germinoma received neither CSI nor whole brain radiation. None of the 18 patients with ventricular radiation relapsed.13 Another study similarly reported three outfield relapses in 39 patients (7%) with exclusive postoperative radiation therapy to the whole supratentorial ventricles.16

Focal radiation therapy alone
In studies that have delivered primarily radiation therapy alone with volumes tailored to the gross disease, focal irradiation limited to the tumor bed is associated with reduced relapse-free survival rates.11,12,16-19 In an analysis of 133 patients receiving focal radiation therapy alone, the local recurrence rate was 6.8%, and the spinal relapse rate was nearly four times higher than for whole brain or whole ventricular radiation (15 patients; 11.3%).5 The total relapse rate was 23.3%.

The addition of chemotherapy cannot meaningfully reduce the recurrence rates. The relapse rate is still between 11% and 66% (median, 16%). Relapses were observed within the ventricles, mainly localized at margin or out of the radiation therapy field.2,20-23 Investigators observed six relapses in 18 patients who received local radiation therapy with a margin of 1.5 cm to the initial gross tumor volume and no relapse in nine patients after ventricular irradiation.24 It should be noted that relapsing disease requires intensive salvage treatments, including CSI and chemotherapy, bearing a high risk for acute complications and severe late effects. Chemotherapy is unable to control subclinical disease within the ventricular system and as a consequence radiation therapy in a combined setting has to include the whole ventricular system. Chemotherapy is, therefore, only causing additional toxicities without a gain in tumor control and survival.

Craniospinal irradiation alone in metastatic disease
Approximately 30% of patients with germinoma have metastatic disease at diagnosis, but reports on disease management are scarce and contain heterogeneous treatment concepts. In an analysis of 129 patients, 51 had multifocal or disseminated disease. Twenty-two received CSI and some received additional chemotherapy. Event-free survival for multifocal or disseminated tumors was inferior to patients with solitary tumors.15 Details, however, were not given.

In the German Maligne Keimzelltumoren (MAKEI) study, 18 of 60 patients had metastatic disease. All received CSI alone without experiencing a treatment failure.25 The data published so far, however, are based on small patient numbers and therefore do not permit a reliable conclusion on optimal treatments. A reduction of treatment volumes even when combined with chemotherapy is most likely associated with an increased relapse rate. Considering the fact that chemotherapy is unable to control intracranial leptomeningeal disease within the ventricular system in localized germinoma, correspondingly subclinical leptomeningeal spread to the entire intracranial and spinal leptomeningeal space in metastatic disease would also be insufficiently treated by chemotherapy. Thus, CSI is necessary and chemotherapy would only add toxicity. Even modern imaging technologies that can detect subtle lesions would not open up the possibility to perform an individualized approach sparing as much normal tissue as possible. Potential areas at risk are, by definition, normal on imaging.

CSI in localized intracranial pure germinoma is the most robust treatment, providing cure in almost all patients. Relapses consisted of initially unrecognized teratoma, initially undetected secreting germ cell tumor, or metachronous tumor within or outside the CNS.25 In completely staged cases target volumes can apparently be safely restricted to whole ventricular radiation therapy plus boost. A prospective randomized trial to show an advantage of whole ventricular irradiation over CSI is biometrically difficult to conduct as this study would require a very large patient number. As with the current philosophy to manage intracranial ependymoma, a straightforward approach to apply this concept in a prospective controlled fashion could be favored.


  1. Bouffet E, Baranzelli MC, Patte C, et al. Br J Cancer. 1999;79:1199-204.
  2. Alapetite C, Ricardi U, Saran F, et al. Med Pediatr Oncol. 2002;39:248.
  3. Kellie SJ, Boyce H, Dunkel IJ, et al. Pediatr Blood Cancer. 2004;43:126-33.
  4. Kortmann RD, Calaminus G, Becker G, et al. Int J Radiat Oncol Biol Phys. 2000;48(Supp 1):204.
  5. Rogers SJ, Mosleh-Shirazi MA, Saran FH. Lancet Oncol. 2005;6:509-19.
  6. Inskip PD, Curtis RE. Int J Cancer. 2007;121:2233-40.
  7. Linstadt D, Wara WM, Edwards MS, et al. Int J Radiat Oncol Biol Phys. 1988;15:291-7.
  8. Shibamoto Y, Abe M, Yamashita J, et al. Int J Radiat Oncol Biol Phys. 1988;15:285-90.
  9. Brada M, Rajan B. Br J Cancer. 1990;61:339-40.
  10. Dattoli MJ, Newall J. Int J Radiat Oncol Biol Phys. 1990;19:429-33.
  11. Wolden SL, Wara WM, Larson DA, et al. Int J Radiat Oncol Biol Phys. 1995;32:943-9.
  12. Shirato H, Nishio M, Sawamura Y, et al. Int J Radiat Oncol Biol Phys. 1997;37:511-5.
  13. Haas-Kogan DA, Missett BT, Wara WM, et al. Int J Radiat Oncol Biol Phys. 2003;56:511-8.
  14. Ogawa K, Shikama N, Toita T, et al. Int J Radiat Oncol Biol Phys. 2004;58:705-13.
  15. Shikama N, Ogawa K, Tanaka S, et al. Cancer. 2005;104:126-34.
  16. Matsutani M, Sano K, Takakura K, et al. J Neurosurg. 1997;86:446-55.
  17. Haddock MG, Schild SE, Scheithauer BW, et al. Int J Radiat Oncol Biol Phys. 1997;38:915-23.
  18. Aoyama H, Shirato H, Yoshida H, et al. Radiother Oncol. 1998;49:55-9.
  19. Shibamoto Y, Sasai K, Oya N, et al. Radiology. 2001;218:452-6.
  20. Cefalo G, Gianni MC, Lombardi F, et al. Med Pediatr Oncol. 1995;25:303.
  21. Allen JC, DaRosso R, Donahue B, et al. Cancer. 1994;74:940-4.
  22. Buckner JC, Peethambaram PP, Smithson WA, et al. J Clin Oncol. 1999;17:933-40.
  23. Nguyen QN, Chang EL, Allen PK, et al. Cancer. 2006;107:2228-36.
  24. Shirato H, Aoyama H, Ikeda J, et al. Int J Radiat Oncol Biol Phys. 2004;60:214-7.
  25. Bamberg M, Kortmann RD, Calaminus G, et al. J Clin Oncol. 1999;17:2585-92.

Current Controversies in Oncology is a forum for the exchange of views on topical issues in the field of oncology. The views and opinions expressed therein are those of the authors alone. They do not necessarily reflect the views or positions of the Editor or of the American Society of Clinical Oncology.

Back to Top