Abstract
Background
This study aims at clarifying the impact of persistent residual lesions following first-line treatment for pediatric medulloblastoma.
Methods
Data on 84 pediatric patients with medulloblastoma and persistent residual lesions on centrally reviewed magnetic resonance imaging (MRI) at the end of first-line therapy were analyzed.
Results
Twenty patients (23.8%) had residual lesions in the tumor bed (R+/M0), 51 (60.7%) had distant lesions (R0/M+) and 13 (15.5%) had both (R+/M+). Overall response to first-line therapy was minor or partial (≥ 25% reduction, minor response [MR]/PR) for 64 (76.2%) and stable disease (SD) for 20 patients (23.8%). Five-year post-primary-treatment progression-free (pptPFS) and overall survival (pptOS) were superior after MR/PR (pptPFS: 62.5 ± 7.0%[MR/PR] vs. 35.9 ± 12.8%[SD], P = .03; pptOS: 79.7 ± 5.9[MR/PR] vs. 55.5 ± 13.9[SD], P = .04). Furthermore, R+/M + was associated with a higher risk for progression (5-year pptPFS: 22.9 ± 17.9%[R+, M+] vs. 72.4 ± 12.0%[R+, M0]; P = .03). Watch-and-wait was pursued in 58 patients, while n = 26 received additional treatments (chemotherapy only, n = 19; surgery only, n = 2; combined, n = 3; valproic acid, n = 2), and their outcomes were not superior to watch-and-wait (5-year pptPFS: 58.5 ± 7.7% vs. 51.6 ± 10.7% P = .71; 5-year pptOS: 76.3 ± 6.9% vs. 69.8 ± 9.7%, P = .74). For the whole cohort, 5-year pptPFS by molecular subgroup (58 cases) were WNT: 100%, SHH: 50.0 ± 35.4%, group-4, 52.5 ± 10.5, group-3 54.2 ± 13.8%; (P = .08).
Conclusions
Overall response and extent of lesions can function as surrogate parameters to predict outcomes in pediatric MB patients with persistent lesions after first-line therapy. Especially in the case of solitary persistent medulloblastoma MRI lesions, additional therapy was not beneficial. Therefore, treatment response, extent/kind of residual lesions and further diagnostic information need consideration for indication of additional treatments for persisting lesions.
Keywords: children, medulloblastoma, MRI, persistent residual disease
Key Points.
Treatment response relates to the outcome of patients with persistent medulloblastoma lesions.
In the case of solitary persistent medulloblastoma magnetic resonance imaging lesion, additional therapy is not beneficial.
Importance of the Study.
The impact of persistent residual disease (ie, non-progressive local residual tumor and/or persistent metastatic lesions and/or lesions of uncertain composition) after completion of first-line treatment in pediatric patients with medulloblastoma is not yet defined. This may result in either over- or undertreatment of these patients. This study included 84 patients with this condition shows that response to first-line therapy is a surrogate parameter for the patient`s outcome and can be used to predict the further clinical course. Further, in this series, additional therapies after first-line therapy had no impact on the risk for disease progression compared to a watch-and-wait strategy. These results will allow clinicians a more evidence-based discussion of options for pediatric patients with persistent magnetic resonance imaging lesions at the end of first-line treatment. Especially, patients with solitary lesions may not benefit from additional treatment.
With an annual incidence of approximately 5 cases per million individuals medulloblastoma (MB) is 1 of the most common malignant central nervous system (CNS) tumors in the pediatric population.1 Current standard therapy includes maximum safe resection, craniospinal irradiation (CSI), and adjuvant chemotherapy. Due to multimodal therapy concepts, prognosis has significantly improved over the last decades with a 5-year overall survival (OS) that exceeds 75% in standard-risk patients.2–4 Nevertheless, the outcomes of patients who present metastatic disease at initial staging are still unsatisfying yielding 5-year progression-free survival (PFS) rates of approximately 60%.5 In addition to conventional clinical parameters such as age at diagnosis, presence of metastases, and extent of resection, recent reports have found that the molecular subtype and the biological profile of the tumor affect prognosis and outcome.6–12 However, the clinical impact of persistent residual disease (ie, local residual tumor or persistent metastatic disease or lesions of uncertain composition) at the end of primary treatment remains unclear.5,13–16 Thus, we retrospectively evaluated a cohort of 84 patients with MB and persistent residual lesions on centrally reviewed magnetic resonance imaging (MRI) at the end of first-line therapy. The primary aim of this study was to evaluate the impact of post-treatment residual disease on prognosis; secondly, risk factors affecting survival in this patient group were analyzed. Finally, the diagnostic work-up and subsequent treatment modalities were comprehensively assessed.
Patients and Methods
Patient Population and Data Assembly
Patients aged ≥4 years at initial diagnosis of MB enrolled in the prospective multi-institutional HIT2000 trial (n = 42), the HIT-2000-Interim registry (n = 11) or the I-HIT-MED Registry (n = 31), who underwent first tumor surgery for MB between 01.01.2000 and 31.12.2019 and had neuroradiological evidence of persistent residual disease (local or distant) at the end of first-line therapy based on central review assessment, were eligible for this study. Patients with radiologically confirmed complete remission (CR) or unequivocal progressive disease at the end of primary treatment were excluded from this study. Central neuroradiological review at initial diagnosis and after completion of primary therapy was a prerequisite for inclusion. Baseline epidemiological and clinical characteristics, tumor- and treatment-related data, as well as outcome parameters, were retrieved from the German HIT-MED databases for this analysis. The prospective HIT2000 trial and the subsequent HIT2000-Interim Registry and I-HIT-MED Registry were approved by the ethical committees of the University of Wuerzburg, Germany and University of Hamburg, Germany (ClinicalTrials. gov identifiers: NCT00303810, NCT02238899, and NCT02417324). All institutions participating in the study received approval from the responsible review boards. Informed consent was obtained from all patients, parents, or legal guardians.
Standard Diagnostic and Therapeutic Procedures
Staging included pre- and postoperative craniospinal MRI and cytological evaluation of cerebrospinal fluid (CSF) obtained by lumbar puncture according to national standards by central review.17 Postoperative imaging was recommended to be performed within 72 hours after initial tumor resection. The extent of surgical resection was determined and classified as < 1.5 cm2 or ≥ 1.5 cm2. Surveillance MRI was done following each treatment element and at the end of first-line therapy whenever CR was not achieved. The extent of the disease was staged according to the Chang criteria (M1, microscopic dissemination into the cerebrospinal fluid [CSF]; M2, macroscopic intracranial metastases; M3, macroscopic spinal metastases; and M4, extraneural metastases).18 Central neuropathological review of tumor specimens was performed at the time of diagnosis at the Brain Tumor Reference Center of the German Society for Neuropathology and Neuroanatomy (DGNN) in Bonn, Germany. For patients with available tumor material, DNA methylation-based CNS tumor classification was performed using the Heidelberg Brain Tumor Classifier v11b48 either retrospectively or as part of initial work-up, eg, via inclusion to the Molecular Neuropathology (MNP) 2.0 Study (PMID: 36928815).
Following maximum safe resection of the primary tumor patients underwent risk-stratified adjuvant radiochemotherapy according to the HIT2000 trial protocol (2000–2011) or national therapy guidelines (2011–present).5,19,20 The HIT2000 protocol and national guidelines provided detailed information for the first-line treatment of children with MB stratified by clinical and more recently also biological risk factors.
Throughout the time period of this study, generally, patients without metastasis at initial diagnosis were to receive direct postoperative craniospinal irradiation (CSI) with boost to the tumor bed, concomitant vincristine intravenously, and subsequent maintenance chemotherapy (HIT-AB4).20 CSI dosage was either 23.4 Gy for standard-risk patients or 35.2 Gy (fractions of 1.8 Gy) for patients with certain high-risk features but not macroscopic metastasis. The hyperfractionated craniospinal dose was 36 Gy, followed by a boost to the whole posterior fossa up to 60 Gy with a further boost up to 68 Gy to the tumor bed in twice daily fractions of 1.0 Gy. In contrast to this, patients with macroscopic metastasis were to receive postoperative chemotherapy including intraventricular methotrexate application, followed by irradiation and maintenance chemotherapy (MET-HIT-AB4). Treatment strategy for patients with isolated microscopic metastasis (M1) changed over time and was analogous to MET-HIT-AB4 until 2008 and direct postoperative irradiation (CSI 35.2 Gy, analogous HIT-AB4) since.5,21 A treatment overview is displayed in Supplementary Figure 1.
Response Assessment and Status at the End of First-Line Treatment
Radiological response assessment via MRI was performed by an experienced neuroradiologist (BB) at the national neuroradiology reference center of the HIT network (University Hospital Wuerzburg, Germany [until 2020] and University Hospital Augsburg, Germany [since 2021]) with response criteria close to the definition later proposed as “RAPNO-MB” (Response Assessment in Pediatric Neuro-Oncology in patients with Medulloblastoma; Supplemental Figure 2). Partial response (PR) and stable disease (SD) were defined as a decrease in tumor volume by more (PR)/less than (SD) 50%.22–24 As the German National Neuroradiological Reference Center also provided a classification using “IMP” (improved; reduction in size between 25% and 49%) for MRIs performed during treatment, additional analyses adding “IMP” to “PR” and defining any reduction in size ≥25% as at least minor response (MR) were performed. This group of patients with a tumor reduction of at least 25% is named “MR/PR” (MR/PR) in the following text. The cutoff for “PR” as proposed by RAPNO (≥ 50%) is used for the respective figures and tables if not otherwise indicated. Response assessment was based on the totality of all lesions rather than on one lesion alone. In this study, the best overall response was ultimately considered for further analyses. According to the location of residual lesions, patients were grouped as R+/M0 (lesion in the tumorbed ≥ 1.5 cm2, no distant lesions), R0/M + (distant lesion, no or < 1.5 cm2 measuring lesion in the tumorbed), and R+/M + (lesion in the tumorbed ≥ 1.5 cm2 and distant lesions). Surveillance imaging was recommended every 3-4 months during the first 2 years after completion of primary treatment.
Statistical Analysis
Statistical analysis was performed using SPSS software (IBM©, version 28). Overall survival was defined from the last day of first-line treatment to death of any cause or to the date of last follow-up (pptOS, post-primary-treatment-OS); PFS was defined as interval starting from the last day of first-line treatment to the date of first progression (pptPFS, post-primary-treatment PFS). pptOS and pptPFS were estimated according to the method of Kaplan and Meier. Log-rank test was used for comparisons of survival in different groups. Potential independent risk factors for pptOS and pptPFS were evaluated by univariate and multivariate Cox regression models. The non-parametric Chi-square test was used to assess significant differences in non-metric parameters. A P-value ≤ .05 was defined to indicate statistical significance without adjustment for multivariable testing. Data were updated as of the last data entry in the database in January 2022.
Results
Patients’ Characteristics and First-Line Treatment
Eighty-four patients from 42 centers in Germany and Austria fulfilled the eligibility criteria for this analysis. Median age at diagnosis was 9.3 years (range 4.03–20.8; Table 1). Most patients were male (n = 68, 81%; female: n = 16, 19%). 91.7% presented with classic histology. The molecular subtype was available for 58 cases (69%) and was group 4 for 38, group 3 for 14, WNT for 4, and SHH for 2 cases (TP53-mutation evaluated and mutated in one patient, not evaluated in the other patient). MYC-amplification status was analyzed for 65 cases (77.4%) and showed MYCN-amplification in one and MYC-amplification in 2 patients. Patients` characteristics are summarized in Table 1.
Table 1.
Demographic and Disease Characteristics of the Study Cohort (n = 84)
| Characteristic | No. of patients | Percentage |
|---|---|---|
| Sex | ||
| Male | 68 | 81.0 % |
| Female | 16 | 19.0 % |
| Histological subtype | ||
| CMB | 77 | 91.7 % |
| LC/A-MB | 6 | 7.1 % |
| DMB | 1 | 1.2 % |
| Molecular subgroup | ||
| WNT | 4 | 4.8 % |
| SHH | 2 | 2.4 % |
| Group 3 Group 4 |
14 38 |
16.7 % 45.1 % |
| Not evaluated | 26 | 31.0 % |
| MYC/N-amplification | ||
| MYC/N-amplification | 1 | 2.5% |
| c-myc-amplification | 2 | 5.1% |
| No amplification | 36 | 92.3% |
| Not evaluated | 45 | 53.6% |
Abbreviations: CMB, classic medulloblastoma (MB); LC/A-MB, large-cell/anaplastic MB; DMB, desmoplastic MB.
All patients underwent surgery as the first treatment element. Early postoperative imaging revealed no evidence of postoperative residual tumor in 21 patients (25%), a small residual tumor measuring < 1.5 cm2 in 18 (21.4%), and a residual tumor ≥ 1.5 cm2 in 45 patients (53.6%). The majority of patients (n = 69/84, 82.1%) had macroscopic metastatic disease at initial diagnosis. Sixty-eight of sixty-nine patients with M2 and/or M3 were treated according to the MET-HIT-AB4 strategy: After 2 cycles of SKK chemotherapy, 55/69 (79.7%) patients received hyperfractionated irradiation whereas 13/69 (18.8%) patients received conventional fractionated radiotherapy. For the remaining patient (1/69, 1.4%), M2 was identified on central neuroradiological review after initiation of irradiation and therefore was treated according to the HIT-AB4 strategy (CSI 35.2 Gy, boost to tumor bed and meningeosis up to 54 Gy). Fifteen patients (15/84, 17.9%) had no evidence of M2/3, but postoperative residual tumor (n = 12/84, 14.2%) or residual tumor with additional M1 (n = 3/84, 3.6%). Twelve of fifteen patients received treatment according to the HIT-AB4 strategy, while 3 patients were treated according to the MET-HIT-AB4 strategy due to R+/M1 disease (n = 2) or large-cell anaplastic histology (n = 1, individual decision). CSI dosage was 23.4 Gy for 5 patients with R+/M0 disease and 35.2 Gy or 40 Gy for 10 patients (hyperfractionated: n = 5; conventional fractionated: n = 5). Seven patients (8.3%) experienced progressive disease during first-line therapy (during pre-irradiation chemotherapy) but responded eventually to irradiation and were therefore included in further analyses. All but 2 patients experienced progression of lesions that were already visible after initial surgery/at first diagnosis. In those 2 patients, a new M2b during induction chemotherapy appeared.
Disease Status at the End of First-Line Treatment
In response to first-line therapy, 64 patients (76.2%) had a PR. In 20 patients (23.8%), the disease remained stable throughout the treatment (Table 2, Figure 1). Twenty patients (23.8%) had residual lesions in the tumor bed only (R+/M0), 51 patients (60.7%) had distant lesions only (R0/M+) and 13 (15.5%) had both distant lesions and lesions in the tumor bed (R+/M+; Table 2). The detailed characteristics of the residual lesions and their evolution during treatment are displayed in Supplementary Table 1 and 2. Three patients underwent various types of positron emission tomography (PET) scan for further diagnostics at the end of first-line treatment. Modalities used were: somatostatin-receptor-PET scan with 68Ga-DOTATOC (showing minimal DOTATOC accumulation), FET-PET (2 FET-positive tumor lesions within caudal vermis), and FDG-PET (no enhanced FDG uptake). Twenty patients had a lumbar puncture at the end of treatment. None of them had positive CSF cytology.
Table 2.
Patients’ Demographic and Disease Characteristics in Relation to the Overall Response to First-Line Treatment (n = 84)
| Characteristic | Overall response during primary treatment = SD (n = 27) | Overall response during primary treatment = PR (n = 57) | P (Chi-Square test) | ||
|---|---|---|---|---|---|
| Sex | 1.0 | ||||
| Female | 5 | (18.5%) | 11 | (19%) | |
| Male | 22 | (81.5%) | 46 | (81%) | |
| Histological diagnosis | .3 | ||||
| CMB | 24 | (88.9%) | 53 | (92%) | |
| LC/A-MB | 2 | (7.4%) | 4 | (8%) | |
| DMB | 1 | (3.7%) | 0 | (0%) | |
| Molecular subtype | .04 | ||||
| WNT | 0 | (0%) | 4 | (7%) | |
| SHH | 1 | (3.7%) | 1 | (1.8%) | |
| Group 3 | 9 | (33.3%) | 5 | (8.8%) | |
| Group 4 | 11 | (40.7%) | 27 | (47.4%) | |
| Not evaluated | 6 | (22.3%) | 20 | (35%) | |
| MYC/N amplification | .4 | ||||
| MYCN-amplification | 0 | (0%) | 1 | (1.8%) | |
| c-myc-amplification | 0 | (0%) | 2 | (3.5%) | |
| No amplification | 14 | (51.8%) | 22 | (38.6%) | |
| Not evaluated | 13 | (48.2%) | 32 | (56.1%) | |
| Findings at the end of primary treatment | .9 | ||||
| R+/M0 | 6 | (22.2%) | 14 | (24.6%) | |
| R0/M+ | 18 | (66.7%) | 34 | (59.6%) | |
| R+/M+ | 3 | (11.1%) | 9 | (15.8%) | |
| Further surgery after the end of primary treatment | .1 | ||||
| Yes | 0 | (0%) | 5 | (8%) | |
| No | 27 | (100%) | 52 | (92%) | |
| Extended treatment after the end of primary treatment | .5 | ||||
| Watch-and-wait | 18 | (66.7%) | 40 | (70.2%) | |
| Chemotherapy only | 8 | (29.6%) | 11 | (19.3%) | |
| Valproate only | 1 | (3.7%) | 1 | (1.8%) | |
| Surgery only | 0 | (0%) | 2 | (3.5%) | |
| Chemotherapy + surgery | 0 | (0%) | 3 | (5.2%) | |
| PD after the end of primary treatment | .5 | ||||
| PD total | 11 | (41%) | 19 | (33.3%) | |
| PD, Local recurrence | 0 | (0%) | 1 | (5.2%) | |
| PD, Metastatic disease | 8 | (73%) | 15 | (79%) | |
| PD, Combined recurrence | 3 | (27%) | 3 | (15.8%) | |
| No PD | 16 | (59%) | 38 | (66.7%) | |
Patients were categorized into 2 groups based on overall response (stable disease, SD vs. partial response, PR; RAPNO classification). All reference assessments made during therapy were considered. In order to detect a correlation between the variables, chi square test was applied. Corresponding P-values are displayed in bold in the right column.
Abbreviations: CMB, classic medulloblastoma (MB); LC/A-MB, large-cell/anaplastic MB; DMB,.
desmoplastic MB; R0, no postoperative residual lesion; R+, postoperative residual lesion ≥ 1.5 cm2; M0,.
no metastatic disease; M + macroscopic metastatic disease; PR, partial response.
Figure 1.
Consort diagram of the patients’ selection process and clinical course, starting with the findings at the end of first-line treatment. For reasons of clarity, study patients were grouped into 3 categories (R+/M0, R0/M+ or R+/M+). Response indicated according to the RAPNO classification. *Inclusion criteria: Age at initial diagnosis was older than 4 years; consent was given; initial surgery between 01.01.2000 and 31.12.2019; not R0/M0 at initial staging; not CR or PD at the end of maintenance therapy; reference findings for initial staging and after the end of primary therapy. Abbreviations: R0, no postoperative residual lesion; R+, postoperative residual lesion ≥ 1.5 cm2; M0, no metastatic disease; M+, macroscopic metastatic disease. CR, complete remission; PR, partial response; SD, stable disease; PD, progressive disease; CT, chemotherapy.
Further Therapy and Clinical Course After Completion of First-Line Treatment
The median follow-up time after the end of primary treatment for survivors (n = 65) was 6.0 years (range: 1.4–16.7 years). After completion of first-line treatment, 54 patients (64.3%) did not show radiological disease activity/progression whereas 30 patients (35.7%) progressed at a median time of 15.8 months (range: 1.3–108 months) after the end of primary therapy. During the observational time of this study, 19 patients (22.6%) died: 17 due to tumor-related causes, 1 patient due to a second malignancy (glioblastoma multiforme, IDH-wild type) and one patient due to intracerebral mass bleeding after bleeding of an epidural seroma. For the whole cohort, 5-year PFS and OS were 61.0 ± 6.0% and 78.4 ± 5.2%. The pptPFS 5 years after completion of first-line treatment was 56.2 ± 6.2% and the 5-year pptOS was 74.3 ± 5.6.
A watch-and-wait strategy was pursued in 58 patients (69%). In 21 of these patients (21/58, 36%) the lesion disappeared resulting in radiological CR during follow-up without administration of any additional therapy or surgery. Twenty-six patients (31%) received subsequent additional treatment procedures (chemotherapy only, n = 19; surgery only, n = 2; surgery and chemotherapy, n = 3; valproate, n = 2). Systemic therapy was mainly based on chemotherapy (temozolomide, n = 19; valproate, n = 2). Surgery included resection in 3 and biopsy in 2 patients. In 2 of these specimens putatively viable tumor cells were detected histopathologically although they did not show progression afterward even though no further therapy was applied (figure 2). In summary, of 84 patients, n = 30 had a relapse or disease progression after the end of first-line therapy. In most of them (27/30), this was detected by surveillance imaging rather than by new or aggravating clinical symptoms. At progression 12/30 patients had exclusively new lesions, 13/30 had a combination of new and preexisting increasing lesions and 5/30 had a growth of preexisting lesions only (Supplementary Table 2). In 19 cases of patients with disease progression after first-line therapy, CSF examination was done. There were no progressions that were exclusively detected by CSF alone. Eight patients had progression on imaging and simultaneous positive CSF cytology. Another 11 patients progressed on imaging, but had negative CSF cytology. For the remaining 11 patients with progression after first-line treatment, CSF cytology was not done. A detailed scheme of outcomes and subsequent additional therapy is given in Figure 1.
Figure 2.
Patients with residual lesions and re-surgery (biopsy or resection) after primary therapy. Abbreviations: R0, no postoperative residual lesion; R+, postoperative residual lesion ≥ 1.5cm2; M0, no metastatic disease; M + macroscopic metastatic disease. CR, complete remission; SD, stable disease.
Differentiation of Viable Tumor and Unspecific Lesion
Resection of the suspect MRI lesion was not possible in most cases to evaluate its composition. Therefore, the following analyses were performed to identify surrogates to better predict if a lesion was more likely to represent active tumor rather than an unspecific radiological alteration or scar.
Patients with large-cell anaplastic MB had a higher risk for disease progression after first-line treatment compared to patients with classical histology (5-year pptPFS: CMB 82.5 ± 5.1% LCAMB 41.7 ± 22.2%, P < .05). All 4 patients with WNT-MB did survive without progression, while both patients with SHH-MB progressed and died. Apart from this observation, patients with group 3 or group 4 MB presented with comparable outcomes (5-year pptPFS WNT: 100%, SHH: 50.0 ± 35.4%, group 4, 52.5 ± 10.5, group 3 54.2 ± 13.8%; P = .08). There was one late relapse in the SHH cohort. The tumor was resected and histological assessment confirmed MB relapse.
In this study, there was no impact on pptPFS and pptOS of the overall response to first-line treatment when using the RAPNO criteria as definition for PR (≥ 50% reduction in tumor size, Figure 3A and B, P = .7 and .6). Yet, pptPFS depended on response to first-line treatment when using a reduction of at least 25% of the tumor size as cutoff for response patients with SD were more likely to experience progression than patients, whose tumors responded minor or partially (Figure 3C and D, 5-year pptPFS: SD 35.9 ± 12.8% vs. MR/PR 62.5 ± 7.0%, P = .03; 5-year pptOS: SD 55.5 ± 13.9% vs. MR/PR 79.7 ± 5.9%, P = .04). In addition, patients with R+/M + at the end of primary therapy were at higher risk to develop progression after the end of first-line therapy compared to patients with R+/M0 (5-year pptPFS: R+/M+. 22.9 ± 17.9% vs. R+/M0 72.4 ± 12.0%, P = .03). 5-year pptPFS was 56.3 ± 7.8% for R0/M+. When comparing nodular to laminar meningeosis, patients with laminar meningeosis had an inferior survival compared to patients with nodular meningeosis only (5-year pptPFS: Nodular only 70.9 ± 9.4% vs. laminar 39.0 ± 11.3, p = 0.03; 5-year pptOS: nodular only 82.2 ± 8.2 vs. laminar 55.3 ± 11.9%, P = .04; Figure 4A and B).
Figure 3.
Post-primary-treatment progression-free survival (pptPFS) and overall survival (pptOS) in years according to overall response during primary treatment. A: pptPFS using RAPNO classification (PR = reduction of tumor size ≥ 50%). B: pptOS using RAPNO classification (PR = reduction of tumor size ≥ 50%). C: pptPFS using German classification (MR/PR = reduction of tumor size ≥ 25%). D: pptOS using German classification (MR/PR = reduction of tumor size ≥ 25%). Abbreviations: PR, partial response; SD, stable disease; MR, minor response.
Figure 4.
Post-primary-treatment progression-free survival (pptPFS) and overall survival (pptOS) in years according to kind of meningeosis and additional therapies after primary treatment. A: pptPFS according to kind of meningeosis. B: pptOS according to kind of meningeosis. C: pptPFS according to additional therapies. D: pptOS according to additional therapies. Abbreviations: CT, chemotherapy.
Yet, outcomes did not differ regarding the location of meningeosis (5-year pptPFS: M2 only 43.9 ± 12.1%; M3 only 57.4 ± 9.9%; M2/M3 combined 49.2 ± 17.7%; P = .8). Likewise the analysis of the contrast enhancement (CE) in residual lesions did not differ between patients who experienced progression during follow-up and patients whose lesions disappeared (5-year pptPFS: All lesions without CE 52.4 ± 7.9% (n = 55) vs. lesions with CE 62.6 ± 10.2% (n = 29), P = .5). Patients whose lesions showed CE that vanished over time (n = 4) had a 5-year pptPFS of 75.0 ± 21.7%.
Moreover, 5-year pptPFS or pptOS for patients receiving any additional therapeutic procedure after the end of first-line treatment was not different compared to patients following a watch-and-wait strategy (5-year pptPFS: Additional procedures 51.6 ± 10.7% vs. watch-and-wait 58.5 ± 7.7%, P = .7; 5-year pptOS: Additional procedures 69.8 ± 9.7% vs. watch-and-wait: 76.3 ± 6.9%, P = .7; Figure 4C and D). Finally, general treatment or specific dosage modifications during first-line treatment, whether per protocol or individual, did not impact pptPFS and OS.
In multivariate regression model including sex, histology, modification of first-line treatment, response to first-line treatment (minor and partial, cutoff ≥ 25% reduction in size), extent of lesions at the end of first-line treatment, location of distant lesions and re-surgery after the end of first-line treatment, the importance of treatment response was confirmed (Hazard ratio 2.3 for SD compared to MR/PR (cutoff ≥ 25% reduction in size), P = .03). Additional multivariate Cox regression including the molecular subgroup and thus excluding the absent molecular cases (n = 58) as well as the same analysis without histological subtype resulted in similar P-values compared to the first model. Other risk factors were well balanced between patients who achieved PR and patients with SD (Table 2; Supplementary Table 3).
Discussion
So far no studies specifically addressed the clinical impact of persistent residual disease or lesions of uncertain composition in patients with MB at the end of first-line therapy. Still, differentiating active lesions from unspecific alterations is of high clinical relevance, as children with tumor progression receiving craniospinal radiotherapy as part of their first-line treatment are considered uncurable.25 In 2014, Fried et al. reported 2 cases of children with initially metastasized medulloblastoma and radiologically visible lesions at the end of treatment that did not show evidence of relapse during a follow-up period of 7 and 13 years.16 They raised the important issue of overtreatment versus undertreatment in these patients. Thus, we analyzed risk factors, clinical course, and additional treatment strategies in a cohort of 84 patients with MB with residual MRI lesions at the end of protocol therapy with the aim of identifying predictive factors differentiating a residual viable tumor from a scar.
First, we aimed at analyzing predictive factors to better decide, whether a residual lesion complies with viable tumor or represents an unspecific alteration (eg, scar) and identified overall response to first-line treatment as most predictive value to assess the risk of progression after the end of primary treatment. Patients with tumor reduction ≥25% of size had a superior 5-year pptPFS and pptOS compared to patients whose disease was stable throughout primary treatment.
Furthermore, patients with both residual metastatic disease and a residual primary tumor had a significantly inferior pptPFS than patients with R+/M0 disease at the end of primary therapy, indicating that an increased tumor burden is a negative predictive factor regarding the risk of viable tumor and consecutive disease progression, although this effect was not observed for pptOS. Additionally, the presence of laminar meningeosis appears to have a negative impact on the further course of affected patients. All this indicates that overall response during primary treatment as well as the extent and kind of residual lesions can help predicting the further clinical course and highlight the importance of central neuroradiological review to ensure a high standard of response evaluation.
Nevertheless, these predictive factors can only give indirect information about the potential risk of a residual lesion representing viable tumor and surgery (either biopsy or resection) still appears to be the most reliable way to assess the viability of post-treatment residual disease, but may not be possible in respect to a reasonable risk for perioperative complications for many patients. Therefore, there is a need for a more refined, less invasive diagnostic repertoire (eg, including liquid biopsy and/or amino acid PET) to define patients at risk for progression. In a previous report by Grosse et al. F-18-Fluorethyltyrosin-PET/CT was found to be a helpful diagnostic adjunct in patients with residual findings on MRI to distinguish vital tumor tissue from unspecific post-therapeutic changes.25 In addition, recent studies in MB patients suggest that liquid biopsy may have the potential to improve early detection of recurrence and identification of non-measurable residual disease. Liu et al. and Afflerbach et al. have shown that circulating tumor DNA (cfDNA) is detectable in CSF of patients with MB with high sensitivity even before structural changes are visible on MRI.26,27
However, in case of unclear residual lesions at the end of primary therapy, these techniques still need to be considered experimental and need further validation in the context of clinical trials before embedding them into clinical routine. The information gained from these experimental investigations may be embedded into clinical management strategies in the future, but has to be prospectively evaluated in the context of clinical trials.
Second, we aimed to evaluate the impact of additional therapies in patients with residual lesions after first-line medulloblastoma treatment.
Our analysis did not identify any difference in pptPFS or pptOS between patients receiving additional therapies and patients managed with a watch-and-wait strategy only. A considerable number of patients from this cohort who achieved a measurable response of at least ≥25% after primary treatment achieved a CR during further follow-up without additional therapies indicating that residual lesions are often non-viable remnants.
Although, this study did not find a benefit using additional therapies, notably, 2 patients treated with valproic acid did not experience progression of their residual lesion. Although low in numbers, this data adds to a growing body of evidence that using histone deacetylase inhibitors (HDACi) such as valproic acid may be a useful approach for the treatment of patients with CNS tumors and, may be used in selected patients with MB and residual lesion.28–30
Apart from these findings, our study has several limitations. Even though this study included 84 patients with persistent findings at the end of primary therapy, the total number of patients is still low. Although histologic subtype and molecular subgroup were found to be related to pptPFS, the low number of patients with WNT- and SHH-MB and LC/A-MB precludes assessing the relation between certain biological risk factors and the degree of response (SD or PR). Furthermore, TP53-mutation was not evaluated in SHH-MB. Although we could not confirm that further therapy contributed to better survival in the cohort as a whole, certain subgroups might have benefited from additional treatment. Despite the large proportion of group 3 and group 4 MB in this cohort, it is yet noteworthy that none of the patients evaluated showed MYC/N amplification. With regard to the histologic subtype, our study displayed a notably high proportion of patients with CMB. In comparison to a previous study by von Bueren with a similar cohort, our analysis shows a significantly higher proportion of patients with CMB (91.7%) versus 80.7% in the reference study.5 In the future, this could be used as a basis for further investigation of the potential correlation between histologic subtype and persistent residual lesion.
Furthermore, effects on pptOS may be overlooked, although median follow-up was 6.0 years and we did not observe many late events. Nevertheless, the high diagnostic standard including mandatory central review of MRI both at initial diagnosis and at the end of primary therapy is a key strength of this study. Therefore, our findings may contribute to improved management strategies in these patients. In conclusion, our results indicate that patients with residual lesions at the end of first-line treatment can be differentiated for their risk of viable, potentially progressing tumors by their neuroradiological response and by the extent and kind of residual lesions. While it seems reasonable to follow a watchful-waiting strategy in patients with a single lesion and minor or PR to primary treatment, patients with multiple lesions or laminar and limited response to primary treatment may have a higher risk for disease progression. Histological assessment from further surgery if possible as well as experimental diagnostic modalities (eg, PET-CT and liquid biopsy) can help to enhance the rationale pro or contra additional therapies. The impact of biological factors needs to be further evaluated in an enlarged cohort.
Supplementary material
Supplementary material is available online at Neuro-Oncology (https://academic.oup.com/neuro-oncology).
Contributor Information
Denise Obrecht-Sturm, Pediatric Hematology and Oncology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Lena Schömig, Pediatric Hematology and Oncology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Martin Mynarek, Mildred Scheel Cancer Career Center HaTriCS4, University Medical Center Hamburg-Eppendorf, Hamburg, Germany; Pediatric Hematology and Oncology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Brigitte Bison, Department of Neuroradiology, University Hospital Augsburg, Augsburg, Germany.
Rudolf Schwarz, Department for Radiotherapy, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Torsten Pietsch, Institute of Neuropathology, Brain Tumor Reference Center of the German Society for Neuropathology and Neuroanatomy (DGNN), University of Bonn, DZNE German Center for Neurodegenerative Diseases, Bonn, Germany.
Stefan M Pfister, Department of Pediatric Hematology and Oncology, Heidelberg University Hospital, Heidelberg, Germany; Division of Pediatric Neurooncology, German Cancer Research Center (DKFZ) and German Consortium for Translational Cancer Research (DKTK), Heidelberg, Germany; Hopp Children’s Cancer Center Heidelberg (KiTZ), Heidelberg University Hospital, Heidelberg, Germany.
Martin Sill, Department of Pediatric Hematology and Oncology, Heidelberg University Hospital, Heidelberg, Germany; Division of Pediatric Neurooncology, German Cancer Research Center (DKFZ) and German Consortium for Translational Cancer Research (DKTK), Heidelberg, Germany; Hopp Children’s Cancer Center Heidelberg (KiTZ), Heidelberg University Hospital, Heidelberg, Germany.
Dominik Sturm, Pediatric Glioma Research Group, German Cancer Research Center (DKFZ), Heidelberg, Germany; Department of Pediatric Hematology and Oncology, Heidelberg University Hospital, Heidelberg, Germany; Hopp Children’s Cancer Center Heidelberg (KiTZ), Heidelberg University Hospital, Heidelberg, Germany.
Felix Sahm, CCU Neuropathology, German Cancer Research Center (DKFZ) and German Consortium for Translational Cancer Research (DKTK), Heidelberg University Hospital, Heidelberg, Germany; Department of Neuropathology, University Heidelberg Heidelberg, Germany; Hopp Children’s Cancer Center Heidelberg (KiTZ), Heidelberg University Hospital, Heidelberg, Germany.
Rolf-Dieter Kortmann, Department of Radiation Oncology, University Hospital Leipzig, Leipzig, Germany.
Nicolas U Gerber, Department of Pediatric Oncology, University Children’s Hospital Zürich, Zürich, Switzerland.
André O von Bueren, Division of Pediatric Hematology and Oncology, Department of Pediatrics, Obstetrics and Gynecology, University Hospital of Geneva, Geneva, Switzerland.
Gudrun Fleischhack, Pediatrics III, Pediatric Oncology and Hematology, University Hospital Essen, Essen, Germany.
Ulrich Schüller, Research Institute Kinderkrebs-Zentrum Hamburg, Hamburg, Germany; Department of Neuropathology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany; Pediatric Hematology and Oncology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Gunther Nussbaumer, Division of Pediatric Hematology and Oncology, Department of Pediatrics and Adolescent Medicine, Medical University of Graz, Graz, Austria.
Martin Benesch, Division of Pediatric Hematology and Oncology, Department of Pediatrics and Adolescent Medicine, Medical University of Graz, Graz, Austria.
Stefan Rutkowski, Pediatric Hematology and Oncology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.
Funding
This study was supported by the German Children’s Cancer Foundation. (Deutsche Kinderkrebsstiftung). The Austrian HIT study center is supported by the Styrian. Children’s Cancer Aid (Steirische Kinderkrebshilfe) [GN, MB].
Conflict of interests statement
The authors declare that they have no conflict of interest.
Authorship statement
Conceptualization: D.O.S., L.S., M.M., and S.R.. Methodology: D.O.S., L.S., M.M., and S.R.. Validation: M.M., M.B., and S.R.. Formal analysis: D.O.S. and L.S.. Investigation: D.O.S., L.S., M.M., B.B., R.S., T.P., S.R., M.B., and F.S.. Resources: D.O.S., L.S., M.M., B.B., R.S., T.P., S.R., F.S., and G.F.. Data curation: D.O.S., L.S., M.M., G.N., M.B., andS.R.. Writing—Original draft: D.O.S., L.S., G.N., M.B., and S.R.. Writing—Review & Editing: all authors. Final approval of the manuscript: all authors.
Data availability
The data analyzed in this study will be made available upon reasonable request.
References
- 1. Northcott PA, Robinson GW, Kratz CP, et al. Medulloblastoma. Nat Rev Dis Primers. 2019;5(1):11. [DOI] [PubMed] [Google Scholar]
- 2. Pollack IF, Agnihotri S, Broniscer A.. Childhood brain tumors: Current management, biological insights, and future directions. J Neurosurg Pediatr. 2019;23(3):261–273. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Gajjar A, Chintagumpala M, Ashley D, et al. Risk-adapted craniospinal radiotherapy followed by high-dose chemotherapy and stem-cell rescue in children with newly diagnosed medulloblastoma (St Jude Medulloblastoma-96): Long-term results from a prospective, multicentre trial. Lancet Oncol. 2006;7(10):813–820. [DOI] [PubMed] [Google Scholar]
- 4. Gajjar A, Robinson GW, Smith KS, et al. Outcomes by clinical and molecular features in children with medulloblastoma treated with risk-adapted therapy: Results of an international phase III trial (SJMB03). J Clin Oncol. 2021;39(7):822–835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Bueren AO, Kortmann R-D, Hoff K, et al. Treatment of children and adolescents with metastatic medulloblastoma and prognostic relevance of clinical and biologic parameters. J Clin Oncol. 2016;34(34):4151–4160. [DOI] [PubMed] [Google Scholar]
- 6. Taylor MD, Northcott PA, Korshunov A, et al. Molecular subgroups of medulloblastoma: The current consensus. Acta Neuropathol. 2012;123(4):465–472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Goschzik T, Schwalbe EC, Hicks D, et al. Prognostic effect of whole chromosomal aberration signatures in standard-risk, non-WNT/non-SHH medulloblastoma: A retrospective, molecular analysis of the HIT-SIOP PNET 4 trial. Lancet Oncol. 2018;19(12):1602–1616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Capper D, Jones DTW, Sill M, et al. DNA methylation-based classification of central nervous system tumours. Nature. 2018;555(7697):469–474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Schwalbe EC, Lindsey JC, Nakjang S, et al. Novel molecular subgroups for clinical classification and outcome prediction in childhood medulloblastoma: A cohort study. Lancet Oncol. 2017;18(7):958–971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Northcott PA, Buchhalter I, Morrissy AS, et al. The whole-genome landscape of medulloblastoma subtypes. Nature. 2017;547(7663):311–317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Cavalli FMG, Remke M, Rampasek L, et al. Intertumoral heterogeneity within medulloblastoma subgroups. Cancer Cell. 2017;31(6):737–754.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Rutkowski S, Von Hoff K, Emser A, et al. Survival and prognostic factors of early childhood medulloblastoma: An international meta-analysis. J Clin Oncol. 2010;28(33):4961–4968. [DOI] [PubMed] [Google Scholar]
- 13. Thompson EM, Hielscher T, Bouffet E, et al. Prognostic value of medulloblastoma extent of resection after accounting for molecular subgroup: A retrospective integrated clinical and molecular analysis. Lancet Oncol. 2016;17(4):484–495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Patel P, Wallace D, Boop FA, et al. Reoperation for medulloblastoma prior to adjuvant therapy. Neurosurgery. 2019;84(5):1050–1058. [DOI] [PubMed] [Google Scholar]
- 15. Thompson EM, Bramall A, Herndon JE, Taylor MD, Ramaswamy V.. The clinical importance of medulloblastoma extent of resection: A systematic review. J Neurooncol. 2018;139(3):523–539. [DOI] [PubMed] [Google Scholar]
- 16. Fried I, Huang A, Bartels U, et al. Chronic residual lesions in metastatic medulloblastoma patients. J Pediatr Hematol Oncol. 2014;36(1):71–75. [DOI] [PubMed] [Google Scholar]
- 17. Hagel C, Sloman V, Mynarek M, et al. Refining M1 stage in medulloblastoma: Criteria for cerebrospinal fluid cytology and implications for improved risk stratification from the HIT-2000 trial. Eur J Cancer. 2022;164(3):30–38. [DOI] [PubMed] [Google Scholar]
- 18. Chang CH, Housepian EM, HerbertC, Jr. An operative staging system and a megavoltage radiotherapeutic technic for cerebellar medulloblastomas. Radiology. 1969;93(6):1351–1359. [DOI] [PubMed] [Google Scholar]
- 19. von Bueren AO, von Hoff K, Pietsch T, et al. Treatment of young children with localized medulloblastoma by chemotherapy alone: Results of the prospective, multicenter trial HIT 2000 confirming the prognostic impact of histology. Neuro Oncol. 2011;13(6):669–679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Lannering B, Rutkowski S, Doz F, et al. Hyperfractionated versus conventional radiotherapy followed by chemotherapy in standard-risk medulloblastoma: Results from the randomized multicenter HIT-SIOP PNET 4 trial. J Clin Oncol. 2012;30(26):3187–3193. [DOI] [PubMed] [Google Scholar]
- 21. Obrecht D, Mynarek M, Hagel C, et al. Clinical and molecular characterization of isolated M1 disease in pediatric medulloblastoma: Experience from the German HIT-MED studies. J Neurooncol. 2022;157(1):37–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Lin NU, Lee EQ, Aoyama H, et al. ; Response Assessment in Neuro-Oncology (RANO) group. Response assessment criteria for brain metastases: Proposal from the RANO group. Lancet Oncol. 2015;16(6):e270–e278. [DOI] [PubMed] [Google Scholar]
- 23. Warren KE, Vezina G, Poussaint TY, et al. Response assessment in medulloblastoma and leptomeningeal seeding tumors: recommendations from the response assessment in pediatric neuro-oncology committee. Neuro Oncol. 2018;20(1):13–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Erker C, Tamrazi B, Poussaint TY, et al. Response assessment in paediatric high-grade glioma: Recommendations from the response assessment in pediatric neuro-oncology (RAPNO) working group. Lancet Oncol. 2020;21(6):e317–e329. [DOI] [PubMed] [Google Scholar]
- 25. Sabel M, Fleischhack G, Tippelt S, et al. ; SIOP-E Brain Tumour Group. Relapse patterns and outcome after relapse in standard risk medulloblastoma: A report from the HIT-SIOP-PNET4 study. J Neurooncol. 2016;129(3):515–524. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Liu APY, Smith KS, Kumar R, et al. Serial assessment of measurable residual disease in medulloblastoma liquid biopsies. Cancer Cell. 2021;39(11):1519–1530.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Afflerbach A-K, Rohrandt C, Brändl B, et al. Classification of brain tumors by nanopore sequencing of cell-free DNA from cerebrospinal fluid. Clin Chem. 2023;70(1):250–260. [DOI] [PubMed] [Google Scholar]
- 28. Häcker S, Karl S, Mader I, et al. Histone deacetylase inhibitors prime medulloblastoma cells for chemotherapy-induced apoptosis by enhancing p53-dependent Bax activation. Oncogene. 2011;30(19):2275–2281. [DOI] [PubMed] [Google Scholar]
- 29. Li X-N, Shu Q, Su JM-F, et al. Valproic acid induces growth arrest, apoptosis, and senescence in medulloblastomas by increasing histone hyperacetylation and regulating expression of p21Cip1, CDK4, and CMYC. Mol Cancer Ther. 2005;4(12):1912–1922. [DOI] [PubMed] [Google Scholar]
- 30. Rahman R, Osteso-Ibanez T, Hirst RA, et al. Histone deacetylase inhibition attenuates cell growth with associated telomerase inhibition in high-grade childhood brain tumor cells. Mol Cancer Ther. 2010;9(9):2568–2581. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data analyzed in this study will be made available upon reasonable request.