Due to the overall rarity of VHL disease and the even scarcer occurrence of CNS hemangioblastomas in pediatric patients, existing screening recommendations are mainly based on small case series and expert opinions. With 99 patients, our study comprises the largest single center cohort. The age of initial manifestation of CNS hemangioblastomas in our study (mean age: 14.4, median: 14.8 years, range: 9–18 years) was higher compared to the results of international VHL cohorts (median age: 13–14 years, range: 6–17 years) [10]. At the time of the study, the screening regimen practiced at our VHL reference center suggested a baseline MRI of the CNS at the age of 14 years [9], which may have contributed to this higher age.
In our study retinal hemangioblastomas (52%) and CNS hemangioblastomas (42%) were the most common manifestation in pediatric VHL patients, this is consistent with previous reports [10]. Our reported frequency distribution of hemangioblastomas across anatomical sections of the CNS corresponds to the literature [3, 8, 21].
Recently updated Danish surveillance guidelines have suggested a baseline MRI scan of the CNS at 10 years of age, followed by MRI of the CNS every second year from the age of 15 years [14]. The American VHL Alliance and a recently published consensus statement by the CNS Hemangioblastoma Subcommittee of the International VHL Surveillance Guidelines Consortium recommend shorter screening intervals in adolescents with biennial MRI scans starting at age 11 [6, 15]. Other groups have advocated a baseline MRI as early as 8 years [16] or at even later start ages of 14–15 years [9, 17, 18].
Determining an optimal age for a baseline MRI requires a balance between early tumor detection on the one hand and the necessity of general anesthesia or psychological effects of screening initiation in early childhood on the other. Based on our findings we have revised the previous Freiburg screening protocol starting MRI imaging at the age of 14 years [9] and recommend starting at the age of 12 years. CNS hemangioblastoma may develop earlier in individual cases than in our cohort, and any neurologic symptoms should lead to timely MRI imaging.
Differing CNS hemangioblastoma burden and age-dependent manifestation rate profiles for the two genotype groups have been reported [5, 11, 22, 23]. The large cohort examined in this study confirmed a significantly higher manifestation rate (hazard ratio = 3.7, 95% confidence interval: 1.9–7.4, p value < 0.0001) and surgery rate (hazard ratio = 3.3, 95% confidence interval: 1.2–8.9, p value = 0.02) for truncating compared with missense mutation carriers (Fig. 3). So far, no specific genetically stratified surveillance protocol has been proposed.
We consider it crucial to advise parents affected by VHL to have their children genetically tested at an early stage, as patients with a truncating variant seem to require special attention. In our study, no gender-specific difference was found regarding manifestation and surgery rates, which contrasts with two recently published international VHL cohort studies reporting that tumors grew significantly faster and new tumors developed at a higher frequency in male patients [11, 24].
As previously reported, CNS hemangioblastomas can follow unpredictable growth pattern influenced by age, sex, genotype, associated cysts, and anatomic location [5, 7, 8, 11, 13, 24]. Overall, 50% of CNS hemangioblastomas show no significant change in size over a long-term follow-up period [11]. In our study, 67% of the tumors demonstrated size stability, which could potentially be attributed to the relatively short average follow-up period of 2 years. Symptom-producing hemangioblastomas are frequently associated with cysts, with the growth rate of cysts typically exceeding that of the solid component [25]. We confirmed that hemangioblastomas with cysts were more likely to require surgery as a result of increased lesion volume. The most frequent preoperative symptoms were related to increased intracranial pressure or cerebellar symptoms such as headache, vertigo, diplopia, emesis, and ataxia (Table 2) which is consistent with data reported in pediatric CNS hemangioblastomas [26].
Symptomatic tumors are universally agreed to require neurosurgical intervention, while the clinical management of asymptomatic but radiologically progressive tumors varies in the literature. The preoperative neurological status is considered an important factor associated with long-term postoperative outcomes in pediatric patients [21]. To prevent the development of irreversible neurological deficits in pediatric patients, the surgical removal of asymptomatic hemangioblastomas with documented radiological progression has been advocated [27]. Our findings support this recommendation, as surgical procedures for asymptomatic tumors could be performed with low morbidity. Two previous studies have reported favorable clinical outcomes for pediatric CNS hemangioblastoma surgery and no cases of local tumor recurrence [26, 27]. This aligns with our observations, however local tumor recurrence could not be ruled out in two cases.
Adolescent patients require special attention as patients 12 to 20 years of age develop more tumors per year than older age groups and cysts grow faster in younger patients [11, 28]. A personalized surveillance plan should be developed considering hemangioblastoma burden, tumor location, tumor size and associated cysts. We recommend MRI examination intervals every (1-) 2 years depending on CNS involvement. Of 29 patients who underwent follow-up MRI examinations for at least one year (mean follow-up 3.8 ± 1.7, range 1.0-7.1 years) after the initial diagnosis of a CNS hemangioblastoma, 22 patients (76%) showed disease progression defined as the manifestation of a new hemangioblastoma (66%) or growth of an existing lesion > 7.5 mm³/year (59%). The mean time until disease progression occurred was 2.0 ± 1.3 years (range 0.3–5.8). The risk of intercurrent CNS hemangioblastoma was reported to be reduced from 7 to 3% when CNS imaging was performed annually instead of every 2 years [29]. Close observation with annually screening intervals may reduce neurological morbidity but must be balanced against potential psychological effects of frequent screening intervals and risks associated with general anesthesia or contrast agent accumulation. For asymptomatic patients with stable or no hemangioblastomas we consider a biennial interval for MRI scans reasonable. Patients with a high tumor burden or progressive hemangioblastoma, especially when associated with cysts, require more frequent screening intervals to weight between watchful waiting and intervention. If new neurological symptoms emerge, an MRI should be scheduled to guide treatment. Patients and even more their parents should be educated on possible cerebellar or spinal symptoms and signs of increased intracranial pressure to raise clinical awareness and enable early intervention. To reduce the psychological burden of frequent examinations and to increase compliance, we recommend a multidisciplinary “one-stop-shop” service for clinical appointments so that all organs potentially affected by VHL disease are examined in one day [30].
LimitationsAs retrospective investigation, this study includes VHL patients followed in our VHL center for whom MRI examinations and surgeries were collected over longer periods of time, so that complete clinical information was not always available. The burden of tumor lesions at the age of the first craniospinal MRI examination depends on the scheduling of the first screening examination, so that it is hardly possible to make a statement about the earlier dynamics of tumor growth. Since the end of observation was reaching adulthood, the follow-up period for the individual lesions was too limited to conduct a comprehensive analysis of tumor growth patterns stratified on anatomic locations. The predominance of the c. 292 T > C mutation among the missense mutation carriers in our cohort may limit the international applicability of our findings. Nevertheless, this is the largest pediatric VHL patient cohort evaluating recommended initiation of routine MRI monitoring of the CNS.
Comments (0)