Tumor predisposition syndromes are hereditary diseases causing a higher risk to develop certain benign or malignant neoplasms in adults and children (1, 2). In adults, the percentage of malignancies attributed to causative genetic alterations is said to be 5-10% (1, 3). In childhood, there is an overlapping but different spectrum of syndromes associated with tumor predisposition, like Li-Fraumeni syndrome, Gorlin syndrome, Fanconi anemia, tuberous sclerosis, neurofibromatosis, Cowden syndrome, APC related adenomatous polyposis, Beckwith-Wiedemann syndrome, etc (1, 4, 5).

Genetic cancer predisposition rates for children have lately been reanalyzed after the discovery of now more than 100 associated genes, and causative germline variants were identified in up to 8.5% of affected individuals (6–8), most of them presenting with an unremarkable family history (8). However, certain entities are more frequently assessed and directed towards genetic testing, making the diagnosis of a hereditary tumor syndrome more likely, e.g. in younger women diagnosed with breast or ovarian cancer or very young children with Wilms or adrenal cortical tumors in (1, 3).

Gorlin, Li-Fraumeni and rhabdoid tumor predisposition syndrome (RTPS) as well as neurofibromatosis type 1 and 2, von Hippel-Lindau syndrome and tuberous sclerosis complex (TSC) are some of the syndromes associated with an increased risk for childhood-onset brain tumors (9, 10). Brain tumors are the most common solid malignancies in childhood with medulloblastomas being the second most frequent entity constituting nearly 20% of all pediatric brain tumors (11–13). Medulloblastomas arise from the cerebellum, vermis or fourth ventricle/posterior fossa and split up in 4 different molecular subgroups: wingless (WNT)-activated (TP53wt), wingless (WNT)-activated (TP53mut), sonic hedgehog (SHH)-activated (TP53wt), MB without WNT/SHH activation (Group 3 or Group 4 (G3/4)) as defined in the 2021 WHO classification (14, 15). They are further characterized by their varying origins, molecular drivers, demographics and clinical outcomes (16, 17).

SHH-activated medulloblastomas (SHH-MB) account for approximately 25% of all medulloblastomas. Almost all SHH-MB contain at least one driver event, most frequently affecting PTCH1, SUFU, TP53 or SMO, KMT2D/2C, HAT, GPR161 and ELP1; with germline mutations in TP53 and ELP1 mostly identified in older pediatric medulloblastoma patients (16, 18–20). The number of damaging germline mutations identified is highest in this subgroup; MYC or MYCN genes are also regularly amplified (16, 21).

SHH-MB occur at two age peaks, in infancy/young childhood and adulthood - with 50% of the affected children being diagnosed before the age of 5 years. At the time of diagnosis 30-40% of patients present with metastatic disease; the 5 year overall survival for this group is only 66% with many long term survivors facing treatment-related neurological adverse effects (22, 23). Median survival time for relapsed disease is still less than 1 year (24).

Patients under the age of 3 will preferably be treated by systemic interval chemotherapy after gross resection of the tumor. Depending on the risk stratification, irradiation can be omitted during initial therapy or delayed until the child is older (or a relapse occurs) (25). Delaying or omitting craniospinal radiotherapy is especially successful in children with non-metastatic disease presenting with desmoplastic or extensive nodular histology which is a strong independent favorable prognostic factor compared to classical MB (26–29). A chemotherapy-only approach (e.g. HIT-SKK regime) is especially favorable in young Gorlin patients where a radiotherapy-sparing treatment option is important to prevent the occurrence of secondary neoplasms like basal cell carcinomas (BCC) (26, 27, 30–34). In line with this, the choice of the primary treatment with the highest possible chance to avoid relapse and consecutively radiotherapy is key (26, 27, 35, 36).

Tumor predisposition syndromes - like PTCH1 or SUFU associated Gorlin and TP53 associated Li-Fraumeni syndrome - affect approximately 7-8% of children with childhood/adolescent cancers and 5-6% of medulloblastoma patients with the highest prevalence of 14-20% for germline mutations in the SHH-MB subgroup (16, 19, 30, 37). Gorlin syndrome is diagnosed at a prevalence of approximately 1:30.000 - 60.000 (38, 39).

The risk for developing a medulloblastoma in Gorlin patients is estimated at 2-5% with a male predominance of approximately 3:1, usually occurring in the first 3 years of life (16, 30). The syndrome was first described in 1960 by Gorlin and Goltz. They initially described a subgroup of patients with basal cell carcinomas, jaw cysts and congenital rib anomalies (40). Subsequently, causative mutations in the genes PTCH1 and SUFU were identified (9, 41). Recently, potentially disease causing heterozygous PTCH2 variants have been identified in patients with a milder Gorlin associated phenotype and controversially discussed but this gene has not yet been included in routine testing for Gorlin syndrome (42, 43). Smoothened (SMO) is another candidate gene frequently harboring somatic mutations in medulloblastoma tumors but also affecting the germline in adult medulloblastoma patients, potentially opening the door for target therapies such as SMO inhibitors (31, 44).

The PTCH1 or SUFU associated Gorlin syndrome follows an autosomal dominant inheritance pattern; up to 80% of the mutations seem to be familial with a sporadic de novo event occurring in 20-30% (38, 39). Offspring of an affected individual will inherit the pathogenic variant in 50%. Penetrance is described to be almost 100% with a highly variable expression (32, 39).

Approximately 60% of Gorlin patients present with typical phenotypic features such as macrocephaly, frontal bossing, coarse facial features, palmar/plantar pits and/or skeletal abnormalities, e.g. of the ribs and vertebrae. Some degree of motor delay is often described, although this is almost always temporary. Global developmental delay is not routinely associated with Gorlin syndrome (33).

Neoplasms in patients with Gorlin syndrome include the typical basal cell nevi/carcinomas (90-100%), jaw keratocysts (90%), cardiac and/or ovarian fibromas (2-20%) and medulloblastomas (5%); most commonly the desmoplastic subtype.

Medulloblastomas occur significantly more often in patients with pathogenic SUFU variants (33%) than in those harboring PTCH1 variants (<2%) (45, 46). Additionally, the risk for radiation-induced meningioma is significantly higher in SUFU mutated patients (47). General life expectancy is not reduced in Gorlin patients (33).

The international HIT-SKK87, HIT-SKK92, HIT2000interim, HIT2000, I-HIT-MED and observation registries were retrospectively screened for patients with suspected or genetically confirmed Gorlin syndrome. A prospective screening for Gorlin syndrome was not part of this study. 2232 patients (0 – 18y) with medulloblastomas were diagnosed between 1987and 2020 and included in one of the registries mentioned above. Out of those 2232 patients, 323 were histologically classified as DMB (0.2y – 17.8y) or MBEN (0.2 – 4.1y) and 1779 as CMB (0.0 – 17.9y) by local pathologists and/or central neuropathological review since 1994 at the Brain Tumor Reference Center of the DGNN at the Institute of Neuropathology, University of Bonn Medical Center, Germany (n=1475). Patients were included in this study if the diagnosis of Gorlin syndrome was genetically confirmed by germline genetic testing and defined as the Gorlin Cohort.

SHH activation was tested by immunohistochemistry or DNA methylation analysis as described previously (48–50).

To form the Comparative Cohort all patients from the existing registries (see above) were screened for age < 3.5 years at the time of diagnosis and SHH activation. To form the M0 Comparative Cohort the patients were screened for M0 status at the time of diagnosis. Gorlin syndrome was not ruled out systematically by genetic testing in all of these patients.

Progression-free survival (PFS) was defined as the time from surgery to first progression (progression or relapse) or date of last follow-up. Overall survival (OS) was defined as the time from surgery to reported date of death or until a certain point in time within the follow-up for a specific patient, e.g. 3 year-OS or 5 year-OS. Survival of patient groups was compared by log-rank test and Kaplan-Meier curves were constructed.

All examinations were carried out on the basis and according to the legal requirements of the revised Declaration of Helsinki of the World Medical Association in 1983. Informed consent was given at study inclusion by the parents or adolescent patients themselves. Corresponding demographic and clinical data were extracted from the existing registry database (see above).

Genetic testing for childhood cancer predisposition syndromes is a rapidly evolving field with significant consequences for the patient himself, the treatment regime and the prognosticated outcome as well as the affected family. Thus, whom to test and when to test will be more relevant now than ever, especially when an oncological diagnosis is first made, and a treatment plan is mapped out. To facilitate these decisions in the future we decided to retrospectively reassess the genetically diagnosed Gorlin patients in our HIT-MED database as well as look at the different patient characteristics such as age at diagnosis, histology, molecular subgroups, residual disease, staging, treatment regimens and clinical outcomes.

As shown in previous studies (55, 56), the age at diagnosis in our cohort was younger than 3.5 years with a range from 0.65 to 3.41 years. Again in line with previous publications (9, 39, 55, 56), all but one patient in our Gorlin cohort presented with medulloblastomas that were histologically classified as DMB (62.5%) or MBEN (37.5%).

In line with recommendations from other groups, nowadays we routinely suggest genetic testing for Gorlin syndrome in patients who are less than 4 years of age at the time of diagnosis and present with an SHH-activated medulloblastoma (9, 37, 57). Retrospectively, in around 1/3 (34/92 patients) of SHH-activated young MBs Gorlin syndrome was reportedly suspected and recommended to the supervising clinician. 16/34 patients (47.1%) were diagnosed with pathogenic PTCH1 (10/16 patients) or SUFU variants (6/16 patients).

Since testing for SHH activation was not available for all 16 Gorlin patients (8/16 patients), we can only assume that SHH activation was present in all of them. Looking at those patients who presented with an SHH activated medulloblastoma aged 3.5 years and younger as well as assuming SHH activation for all our Gorlin patients, approximately 17% of them would have tested positive for a pathogenic PTCH1 or SUFU mutation.

Since genetic testing was not conducted in all patients in our cohort, we unfortunately cannot evaluate the cumulative incidence of Gorlin syndrome. Taking into account that 16 out of 2187 patients from our cohort of patients with medulloblastomas were genetically diagnosed with Gorlin syndrome, we can only attest for these 0.7%. Judging by the low number of genetic analysis conducted within this cohort, e.g. only 4/16 cases (25%) were screened for germline variants by the MNP 2.0 study (36) in our cohort, and the fact that some medulloblastoma associated genes have only recently been described and not yet added to the standard workup for patients with SHH activated MB (19), we hypothesize that the number of undetected cases is much higher.

HIT-MED Guidance currently recommends SKK chemotherapy as initial therapy for SHH activated, low risk medulloblastomas such as DMB and MBEN in children aged 0-4/5 years (see current HIT-MED Guidance) (26, 27). SKK chemotherapy was administered in all patients initially treated in Germany (15/16) and started in the remaining patient after relocating treatment to Germany. 10/16 patients (62.5%) achieved complete remission after completing their initial treatment regimen; the 10y-OS for all Gorlin patients was 93.3% ± 6.4%. This is in line with previously described 10y-OS from our group of 82% ± 12% for Gorlin patients (55).

Radiotherapy was not administered as part of the initial therapy in any of our Gorlin patients (0/16). This is in concordance with many radiotherapy-sparing treatment regimens for young children with medulloblastomas, potentially avoiding the devastating long-term neurological effects (26–28, 35). It was however administered later on in 2 patients (12.5%); once as part of a relapse regime and once in a patient with stable residual disease after initial R+ resection. One of the two relapsed patients terminated radiotherapy after the diagnosis of Gorlin syndrome and in one other patient radiotherapy was not administered at all during disease progression because of the genetic diagnosis (see Figure 2 ).

Overall 5/16 patients (31.3%) developed additional neoplasms (see Figure 3 ): they comprised of basal cell carcinomas, odontogenic/jaw cysts, ovarian fibromas and a meningioma. These tumors mostly developed in patients who did not receive radiation at any point. While we do not know for certain when some of these tumors appeared, some were reported before the start of the treatment of the medulloblastoma. These findings question reports that postulate a higher risk for secondary malignancies in Gorlin patients after radiotherapy and suggests a generally increased risk for such entities due to the genetic diagnosis of Gorlin syndrome and/or – possibly even more likely - after the administration of chemotherapy or radiotherapy in these mutation-prone patients (41, 58–60). Because Gorlin patients present with a generally increased risk for neoplasms of up to 100% (33), we decided to title the secondary neoplasms documented in this cohort Additional Neoplasms for accuracy (see Results).

Ideally, genetic counseling should be offered to all families of young children with DMB, MBEN or SHH-activated medulloblastoma directly after diagnosis and germline testing performed in a laboratory certified for these analyses. However, not all patients are offered this opportunity or follow these recommendations. When including patients in studies like the MNP study (36), somatic variants in tumor DNA might be identified like in our patients. These patients should then be referred to undergo genetic counseling and be tested for the germline variant. Usually in Gorlin patients, these variants will be confirmed in lymphocyte DNA, like in our patients. However, there are cases where somatic SUFU or PTCH1 variants were identified in tumor DNA but could not be confirmed in the germline. If these variants appear in DNA from multiple neoplasms of the same patient or the patient presents with additional features of Gorlin syndrome, mosaicism should be considered (34, 61).

Overall survival within the Gorlin patients varied depending on mutational status. Potentially at least partially owed to the shorter follow up period, SUFU mutated patients had a 10y-OS of 100% while PTCH1 mutated patients had a 10y-OS of 90%. The difference between these two groups was not statistically significant (p=0.414). However, the 5y-PFS in the SUFU mutated cohort was 41.7% ± 22.2%, while the 5y-PFS in the PTCH1 mutated cohort was 88.9% ± 10.5% - even though this difference again was not statistically significant (p=0.139). This is in concordance with previous studies hinting at a reduced PFS and OS for SUFU mutated patients compared to PTCH1 mutated patients or SHH activated DMB/MBEN medulloblastoma patients in general (45, 57). We hypothesize that a bigger sample size and/or longer observational period might have yielded a statistically significant difference with a less favorable outcome for SUFU mutated patients.

To assess this further, we compared OS and PFS of Gorlin patients to SHH activated, MBEN/DMB, younger than <3.5 years at the time of diagnosis from the registries: Neither Gorlin vs. the Non Gorlin/Comparative Cohort (p=0.228), nor SUFU vs. PTCH1 + Non Gorlin cohort (p=0.122) nor Comparative/Non-Gorlin cohort vs. PTCH1 vs. SUFU (p=0.302) yielded significantly different results in terms of PFS and OS.

To exclude a bias owed to metastatic disease in some of the Gorlin patients in this cohort, we created a subgroup of Gorlin patients with M0 status (Gorlin M0 cohort, n = 13). To evaluate potential differences in PFS and OS for this cohort we compared these patients to the recently published SHH-activated, M0 infant medulloblastoma cohort by Mynarek et al. who also received SKK chemotherapy (Infant MB) (28). Again, there was no significant difference in terms of OS (p=0.812) or PFS (p=0.746) between these cohorts. Also, comparing non SUFU and SUFU patients in these cohorts alone, there was not significant difference between 3y-OS (92.1% ± 3.8% vs. 100%; p=0.623) or 3y-PFS (86.2% ± 4.9% vs. 75% ± 21.7%; p=0.337). After exclusion of the potential confounding variable that is metastatic disease, these findings confirm that SUFU mutated patients tend to present with more events over the course of time (reduced EFS/PFS), but larger numbers and longer follow up would be needed to assess for statistically significant differences between the SUFU and PTCH1 versus non Gorlin MB patients.

When assessing PFS and OS we noticed the significant disparity in follow up periods for the PTCH1 and SUFU mutated cohort (12.1 years vs. 2.58 years). We can only assume, that a longer follow up period especially for the SUFU mutated cohort might have altered some of the reported differences in PFS and OS to a statistically significant level. As to why the follow up for this subgroup was so much shorter, we cannot exclude the possibility that SUFU associated medulloblastomas represent a new tumor entity that recently rose in numbers. But more likely we assume that this is owed to the fact that SUFU mutations were first described in 1999 and only a few years ago added to the list of MB/Gorlin associated genes (62). Because of this timeline, a comparative prospective analysis of PTCH1 vs. SUFU associated Gorlin patients would potentially only be valid from the year 2000 on. More contradictory still, our SUFU patients present with a median follow up of 2.58 years only with most SUFU mutations identified within the last 0-4 years, allocating the least recent SUFU diagnosis in our cohort to the year 2014 (4 I-HIT-MED patients, 2 patients from the Interim Registry). We hypothesize that a retrospective genetic analysis of older PTCH1 negative, SHH activated infant MBs would potentially reveal more SUFU associated Gorlin patients as demonstrated by Smith et al. (2014) (47). Furthermore, recently PTCH2 and possibly also SMO mutations have been described as a potential cause for Gorlin syndrome associated medulloblastoma (43, 63, 64). These genes have not (yet) been implemented in routine genetic testing strategies for patients with the suspected diagnosis of Gorlin syndrome and might (partially) close the diagnostic gap in young children with SHH-activated medulloblastomas lacking germline PTCH1 or SUFU mutations.

In line with previous reports, 5 patients in our cohort presented with motor and/or global developmental delay during the follow-up period (27, 60). While temporary motor delay is often reported in children with Gorlin syndrome, persisting global developmental delay is not a common finding (31). There are few affected cases in the literature, which may be caused by more complex genetic alterations, such as the 9q22.3 microdeletion encompassing the PTCH1 gene (65). However, point mutations have not commonly been associated with global developmental delay/intellectual disability (ID) (33). While developmental delay after the treatment of medullobastoma in children with Gorlin syndrome might be a secondary effect of the treatment itself, the occurrence of developmental delay before the diagnosis of the medulloblastoma is uncommon.

Interestingly, two additional dysmorphic Gorlin patients in our cohort – one harboring a PTCH1 and the other a SUFU mutation - presented with developmental delay prior to the diagnosis of the medulloblastoma/the beginning of treatment. Unfortunately, follow-up neuropsychological evaluation was not available for these 2 patients, but this finding might illustrate the need to monitor the global development of Gorlin patients closely, notwithstanding the diagnosis/treatment of a medulloblastoma.

All SHH activated medulloblastoma patients younger than 4 years of age at diagnosis – especially DMB and MBEN - should be evaluated for Gorlin syndrome and systematically undergo specific genetic testing. Taking our data and previously published data on Gorlin patients into account, we can only assume that there are many Gorlin patients out there who escaped diagnosis. Since this diagnosis affects treatment, clinical management, familial planning and strongly influences the outcome, the diagnosis of Gorlin syndrome in a patient with medulloblastoma should be made as early as possible. Effective chemotherapeutic treatment strategies aiming to avoid radiotherapy during primary treatment are available, but the optimal regimen for Gorlin patients needs to be further investigated. More retrospective and prospective international studies to assess treatment, long-term survival and secondary neoplasms of the PTCH1 and SUFU mutated Gorlin subgroups are warranted.

The data analyzed in this study is subject to the following licenses/restrictions: Privacy restrictions (patient names, family pedigree, clinical course, etc) in HIT-MED database. Requests to access these datasets should be directed to KK, k.kloth-stachnau@uke.de or katja.kloth@gmail.com.

Ethical review and approval were not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

KK, SR, and MM contributed to conception and design of the study. KK, MM, DO, and SR accessed the database and generated the dataset. KK performed the statistical analysis and wrote the first draft of the manuscript. SR supervised the writing of the manuscript and the design of the study. KK, MM, DO, BB, DS, TP, MW-M, and SR provided relevant data. All authors contributed to manuscript revision, read, and approved the submitted version.

The HIT-MED trial office and the Neuroradiological Reference Center are supported by Deutsche Kinderkrebsstiftung.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.