Time to dismiss boost? Outcomes of children with localized and metastatic germinoma

To determine long-term outcomes of a cohort of children with germinoma treated with chemotherapy and radiation therapy without primary tumor boost even in the absence of complete response to chemotherapy This retrospective study analyzed the outcome of patients with germinoma consecutively diagnosed and treated at a tertiary care center from January 2000 to December 2021. MRIs were reviewed by two radiologists, blinded to patient data. Tumor location at diagnosis, tumor response to chemotherapy and at completion of radiation therapy and site of relapse were assessed. Tumor response was assessed radiologically by determining the tumor size and response on diffusion-weighted imaging, in addition to biochemical, cytological parameters and neurological status. Of 46 pediatric germinoma patients, 29 children (14 male; median age 12.8 years) received no primary tumor boost. Median follow-up was 63 months (range 9–187 months). Twenty-five children had localized disease and tumor location was suprasellar (n = 11), pineal (n = 10), bifocal (n = 3) and basal ganglia (n = 1) while 4 children had metastatic disease at presentation. All patients completed multi-agent chemotherapy followed by either ventricular irradiation (VI) (23.4 Gy) (n = 23), whole brain (WBI) (23.4 Gy) (n = 5) or craniospinal radiation (CSI) (23.4 Gy) (n = 1). Two children, who had localized disease at presentation and received VI after chemotherapy, relapsed 9 months and 32 months after completion of treatment respectively. No patient had a local relapse. Location of relapse was distant, outside (n = 1) and out- and inside (n = 1) the irradiation field. Five-year progression free survival (PFS) was 91% and overall survival (OS) was 100%. In this case series, excellent 5-year PFS and OS rates were achieved with chemotherapy followed by radiation therapy of 23.4 Gy delivered without primary tumor boost. No local relapse was observed despite omitting primary tumor boost in patients with localized and metastatic germinoma.


Background
With the established radio-sensitivity of germinomas, radiation therapy alone has been the standard of care for patients with germinomas for many years and remains the standard of care in most adult institutions. The German Cooperative Prospective Trials MAKEI 83/86/89 demonstrated 5-year relapse-free and overall survival rates of 91.0% and 93.7% respectively, with only craniospinal irradiation (CSI) of 30-36 Gy with a primary tumor boost up to 45-50 Gy [1]. While effective, dose and field of irradiation have been raising concerns of significant endocrine and neurocognitive toxicities. Subsequent endeavors culminated in using induction chemotherapy to permit reduction in both the radiotherapy field and irradiation dose while preserving a favorable 1 3 outcome. International trials of the Japanese Study Group [2], SFOP [3] and SIOP 96 [4], using induction chemotherapy followed by focal radiotherapy to the primary tumor bed observed increased relapse rates within the ventricular system which prompted the adoption of ventricular irradiation (VI) post chemotherapy in ensuing trials.
Results from ACNS1123 [5] and SIOP GCT II [6] emphasized that complete tumor response to induction chemotherapy allows either VI dose reduction to 18 Gy with tumor bed boost to a total of 30 Gy or VI to 24 Gy with omission of boost respectively. While the presence of residual lesions at the end of treatment did not demonstrate a negative impact on progression-free survival (PFS) [4], it remains unclear whether residual disease at the end of chemotherapy necessitates higher irradiation doses.

Study objectives
The primary objective of our study was to retrospectively examine the long-term outcomes of germinoma patients who received chemotherapy and radiation therapy without primary tumor boost even in the absence of complete response to chemotherapy.

Patients and methods
The study was approved by the institutional research ethics board and informed consent was waived due to the retrospective nature of the study.
All patients with newly diagnosed germinoma treated at the Hospital for Sick Children, Toronto, Canada from January 2000 to December 2021 were included ( Table 1). The study sample included all 24 germinoma patients treated at the Hospital for Sick Children from January 2000 to December 2013 described and published by Cheng S et al. [7]. The latest clinical status of these patients was updated and included in this analysis.
Demographics, disease characteristics and treatment of germinoma patients who received no primary tumor boost were further analyzed to include treatment response after chemotherapy and radiotherapy, and characteristics of relapse if any. The decision to omit boost was guided by data presented by SIOP and Japanese working groups at international conferences as well as after a thorough literature review and implemented as an institutional approach consistently since 2006.
For the focused cohort of patients who received no primary tumor boost, photon VI, whole brain or CSI, 23.4 Gy in 13 fractions depending on the primary tumor location, was administered after chemotherapy. Chemotherapy consisted of either carboplatin and etoposide or carboplatin/ etoposide/ifosfamide. MRI re-review was performed by two independent radiologists, blinded to the initial evaluation. Tumor location at diagnosis, tumor response to chemotherapy and at completion of radiation therapy as well as site of relapse were documented. Based on the consensus recommendation from the European Society for Pediatric Oncology Brain Tumor Group and North American Children's Oncology Group [8], tumor response was assessed radiologically by determining the tumor size and response on diffusion-weighted imaging, in addition to biochemical, cytological parameters and neurological status.
Clinical data are expressed as medians and as ranges, where appropriate. PFS and overall survival (OS) were evaluated using the Kaplan-Meier method.
Serum βHCG was elevated in 4 (13.8%) patients, ranging from 5 to 30 IU/L. Eight patients (27.6%) had elevated CSF βHCG. Of note, two patients with obstructive hydrocephalus and ventricular CSF βHCG had levels of 96 and 113 IU/L, whereas lumbar CSF βHCG of the remaining 6 patients was in the range of 5-32 IU/L. Staging was medically contraindicated due to obstructive hydrocephalus in one patient with bifocal germinoma. All patients with elevated serum βHCG (mean 15 IU/L) demonstrated higher levels of CSF βHCG (mean 46.5 IU/L). AFP level in serum and CSF was ≤ 8 ng/ ml and negative, respectively, in all patients. Six of the 29 patients did not have biopsy/ surgical resection at diagnosis and were enrolled based on classical clinical and radiological characteristics of either bifocal tumor (n = 3), elevated βHCG (n = 2), or both (n = 1). Twenty-two patients had a diagnostic biopsy, one patient underwent transnasal debulking of her suprasellar lesion; no patient had second look surgery.
Twenty-five children had localized disease. The most common tumor locations were suprasellar in 11 (37.9%) and pineal in 10 (34.5%). Three (10.3%) patients had bifocal tumors and one (3%) had basal ganglia lesion. Four (16.0%) children had metastatic disease at presentation. Of those, one patient had pineal germinoma with 4th ventricular metastasis. One patient had suprasellar and basal ganglia involvement and two patients had bifocal disease; one with metastases within the lateral ventricles and bilateral internal auditory canal and the other with spinal metastases.
All patients received multi-agent chemotherapy prior to radiotherapy. Eighteen (62.0%) and 4 (13.8%) patients received 4 and 3 cycles of carboplatin/etoposide, respectively. Six (20.7%) patients were treated with 4 cycles of carboplatin/etoposide alternating with ifosfamide/etoposide. One (3.5%) patient with an underlying rare metabolic disease (=CACT deficiency) developed anaphylaxis to etoposide/phosphorylated etoposide and proceeded subsequently to 4 cycles of carboplatin and weekly vinblastine.
After induction chemotherapy, complete response and partial response were achieved in 24 (82.8%) and 3 (10.3%) patients respectively. Of the 3 patients with partial response, one patient with localized disease had pineal residual disease of 1.3 cm × 0.8 cm × 1.0 cm (from 2.0 cm × 1.8 cm × 1.9 cm) on the antero-posterior, transverse and cranio-caudal dimensions (AP x TR x CC). The other two patients, both with metastatic bifocal germinoma, achieved complete response at their metastatic sites but not at their primary sites. One had residual disease in the pineal gland measuring 0.9 cm × 1.5 cm × 1.1 cm (decreased from 3.8 cm × 4.7 cm × 4.6 cm at presentation). The other had residual disease in the suprasellar region of 0.9 cm × 1.1 cm × 0.7 cm (decreased from 4.6 cm × 3.2 cm × 3.7 cm at presentation). Assessment of response was indeterminable in two patients (6.9%) with basal ganglia germinoma (localized in one and disseminated in another) as inherent in this location.
In all patients achieving less than complete response after chemotherapy (including the two patients with unevaluable basal ganglia disease), the residual radiological changes uniformly demonstrated no further improvement after radiotherapy. None of these patients had a relapse of the disease. Two children, who had localized suprasellar disease at presentation and received VI after achieving complete response post chemotherapy, relapsed 15 and 36 months from initial diagnosis, respectively. None of them had a local relapse. In these two patients, the locations of distant relapse were outside (n = 1, posterior fossa) (details provided in Cheng et al. [7]) and outside and inside (n = 1, pineal, left hippocampus and posterior mesial temporal region) the radiation field (Table 3). Intriguingly, the latter patient underwent transnasal debulking at diagnosis and-due to postoperative complications and CSF leakthe initiation of chemotherapy was delayed by 6 weeks.
In this study, cohort of patients who received no primary tumor boost, we observed a 5-year PFS of 91% and OS of 100%. This was comparable to the cohort of patients who received primary tumor boost as part of radiotherapy in addition to chemotherapy, which achieved a 5-year PFS and OS of 100% respectively. Collectively, the 5-year PFS and OS of germinoma patients treated from January 2000 to December 2021 at the Hospital for Sick Children were 95% and 100% respectively (Fig. 1).

Discussion
Our retrospective single-center study demonstrated excellent outcomes in germinoma patients treated with chemotherapy followed by radiation doses of 23.4 Gy (VI, WB, CSI depending on location and dissemination) without boost. Omitting primary tumor boost in patients with localized and metastatic germinoma even in the absence of complete response to chemotherapy, led to excellent 5-year PFS and OS of 91% and 100% in our patient sample. While late failures after 5 years are described, those are exceptionally rare [3]. Our results are comparable to the outcome of germinoma patients treated with primary tumor boost [3][4][5]7]. The reduction of late effects of irradiation has been an ongoing effort to improve quality of life of patients, in line with the Delphi consensus statement (2nd statement) on management of intracranial germ cell tumor [9]. The chemo-sensitivity of germinoma is well documented with response rates as high as 92% [5]. However, attempts of radiation-sparing treatment with chemotherapy only resulted in high recurrence rates [10,11]. Hence, combination of chemotherapy followed by radiation therapy is the current international standard employed by pediatric oncologists [2][3][4].
Germinoma treatment continues to evolve with experiences from SIOP GCT II [6] and ACNS1123 [5] trials using response-based radiotherapy. This implies that the status of residual disease after induction chemotherapy emerges as a key factor in stratifying radiotherapy dose, especially when surgical resection, the other strategy for local control of residual disease, has understandably high risk of morbidity involving precarious yet prevalent areas such as the pituitary area, pineal gland and less commonly the basal ganglia.
The presence of residual lesions at the end of treatment is considered an adverse prognostic factor for most malignant pediatric brain tumors. For germinoma, however, previous experience from the SIOP GCT 96 study did not demonstrate a negative impact on PFS with presence of residual lesions at completion of therapy [4]. Furthermore, in the ACNS1123 trial, germinoma patients with residual disease  Fig. 1 Relapse-free survival curve of 46 patients with germinoma more than 1.5 cm in diameter were mandated to undergo surgical resection prior to radiotherapy to remain on study. Pathology revealed either fibrosis, non-viable tumor or mature teratoma in all patients who underwent (second) surgery for this reason [5]. Therefore, based on SIOP GCT II results concluding VI with 24 Gy is sufficient for disease control in localized germinoma with complete response [6], the role of radiotherapy dose more than 24 Gy or providing additional primary tumor boost to residual tumor, which does not necessarily contain viable tumor tissue, requires re-examination. The long-term side effects of irradiation on neurocognition and quality of life in children are well established. Notably, patients with CNS germ cell tumors receiving radiotherapy in a dose range of 25.8-39.9 Gy suffered decline over time in working memory, processing speed, and visual memory, reported by Mabbott and colleagues [12]. Similarly, Wilkening et al. revealed high incidence of measurable memory deficits and amnesia in patients with CNS germ cell tumors [13]. It is expected that the pending results on neurocognition evaluated during the ACNS1123 study will further elucidate radiation dose dependent challenges. Future mitigation techniques that should be explored include reduction of the radiation prescription on clinical trials for germ cell tumors and use of newer radiation techniques to deliver ventricular radiotherapy, such as proton therapy [14].
In our study, the assessment of treatment response was based on the criteria from the consensus recommendation of the European Society for Pediatric Oncology Brain Tumor Group and North American Children's Oncology Group [8]. While assessment of treatment response of non-measurable metastatic disease has rather clear categorical description (complete disappearance, presence or absence of progression), measuring radiological response in the target lesion/s in mainly the pituitary and pineal region is largely dependent on the maximum dimension of the lesion size and degree of restricted diffusion. Focusing on the measurement specifically, complete response and partial response can differ literally by just millimeters. Therefore, interrater discrepancy in measuring the residual disease is predictably a challenge. In addition, bias is likely, especially when the rater recording radiological measurement is cognizant of the extra millimeter that may change the response category and ultimately the radiotherapy dose (as well as the possible late effect from the dose difference) based on the current response-based radiotherapy strategy. In addition, although pure germinoma is typically a solid lesion, cystic components can arise during response to chemotherapy. The evaluation of response excludes cystic components, which adds to the challenge of measurement of the residual disease [8].
Assessment of basal ganglia disease poses different challenges. In our study, there were two patients with basal ganglia involvement and treatment response was indeterminable.
In terms of lesion size, unlike pituitary and pineal lesions, there is no stipulated measurement provided in the guideline regarding basal ganglia lesions, complicating the inherent measuring difficulty of lesions within this area. Furthermore, in our patients with basal ganglia lesions not demonstrating restricted diffusion at diagnosis, the radiological response criterion based on restricted diffusion was not applicable. The indistinct radiological characteristics of basal ganglia germinoma and variable radiological features such as tumor composition, enhancement and diffusion restriction pattern [8,15] emphasize another potential concerning area of interrater reliability with impactful consequences. There is an emergent need to refine the guideline on radiological response by supplementing the criteria with additional reliable and measurable parameters [15].
Our study demonstrated comparable outcomes with a uniform dose of radiotherapy disregarding the presence of complete remission in germinoma. The radiological assessment of treatment response in germinoma could potentially focus on absence of diffusion restriction to better distinguish between residual disease and treatment related changes [8].

Conclusion
Our retrospective single-center analysis demonstrated excellent 5-year PFS and OS in patients with germinoma receiving chemotherapy followed by radiation therapy delivered without primary tumor boost even in incomplete response. Despite omission of the primary tumor boost, our study demonstrated absence of local relapse in patients with localized and metastatic germinoma. Our study confirmed the feasibility of avoiding irradiation boost in the context of 23.4 Gy VI, paving the way of minimizing longterm sequelae of irradiation. Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Conflict of interest
The authors have not disclosed any competing interests.