Discussion
Our study sought to describe the epidemiological characteristics and provide survival outcome data of children with medulloblastoma, age 3-18 years in a single, public tertiary-care referral center located in Rio de Janeiro/Brazil.
Five-year event-free survival of this cohort was 58.4% and the overall survival was 59.1%, for patients >3-18 years at diagnosis, and 5-year OS 69.4% for SR and 53,8% for HR patients. In a retrospective study from 1983 to 2001 at the same institution with 101 patients under 18 years, the 5-year OS was 53% and the 5-year Disease Free Survival (DFS) rate was 40%(21). According to staging, our cohort had 5-year OS of 52.9% for M+ and 71.0% for M0 (p=0.019). In previous Brazilian studies, patients M+ had 5-year OS of 43.7% and M0 had 5-year OS of 73.8%(21) and 5-year OS 59% for SR patients and 5-year OS 24% for HR (22).
High-income countries generally have better survival outcomes compared to lower income countries. The COGA9961 trial, conducted in the United States, had 5-year EFS of 81% and OS 87% for standard-risk patients(20). Other reports from Canada(23) (OS: 69.2%) and United Kingdom(24) (OS: 73.4%) show similar survival outcomes. However, some high-income countries show lower survival rates, such as Spain(25,26) with reported 5-year OS of 55% and EFS of 46%, Singapore(27) with 5-year OS of 51.5% and EFS of 44.5%, and Norway(28) with 5-year OS of 62%. Some upper middle-income countries have reported variable outcomes, such as Taiwan(29) where 5-year OS has been reported 65.9% in one institution and 50% in another institution(30).
The difference in survival among high income (HIC) and low-middle income countries (LMIC) is multifactorial. Late diagnosis with advanced disease presentation, coexisting debilitating conditions (such as malnutrition), treatment abandonment, and inefficient health care systems are some of the barriers of care in pediatric cancer in LMIC(31).
Treatment abandonment has been considered one of the most important causes of cancer treatment failure in LMIC(38) and is defined as the failure to start or complete medically-indicated, possibly curative treatment causing unnecessary mortality and morbidity(39). The causes of treatment abandonment are multifactorial – financial, social, political, health care availability – and are frequently beyond the family’s possibility to control. The duration to define abandonment is ≥4 consecutive weeks of missed therapy. Since patient refusals of treatment without resuming therapy are not always recorded, treatment abandonment has been historically underreported and misclassified and the reported rates are variable, ranging from 0 to 74.5%. At INCA, the report from 1139 patients between 2012-2017 was 1.66% for all pediatric solid tumors and 0.7% for pediatric brain tumors(20). The main reasons for abandonment were lack of understanding the need for therapy, followed by transportation and financial difficulties. In this present study there was 2 cases of abandonment of treatment after radiation therapy (1.6%). Both patients are alive and resumed the follow-up without chemotherapy, confirming their result that abandonment was not associated with increased mortality.
Survival was also assessed by the distance from home to the tertiary cancer center. Patients who lived more than 40km from INCA fared better than those who lived closer, (5-year OS: 68,2% vs 52%; p=0.032). This counterintutive conclusion can be possibly explained by socioeconomic conditions that Brazilian patients face during their treatment. Patients who live far from INCA can stay at housing provided by Ronald McDonald House Charities, which is near the hospital and has transportation at any time of the day and in case of emergencies (such as fever). So even the poorest patients hosted there can access the hospital promptly when needed, reducing the time to get medical support or diminishing absenteeism. Patients often complain about lack of transportation to acess the hospital for outpatient consultations or for emergencies, and while the government provides some resources for patient transportation, it is not widely distributed or utilized.
The use of initial chemotherapy after surgery and delayed radiation therapy has shown worse outcomes when compared to upfront radiation therapy(32). In this cohort, a group of patients (n=12) received pre- irradiation chemotherapy by week 15. Eighty-three patients received upfront radiation therapy post surgery according to the COGA9961 protocol; this group had increased OS when compared to the chemotherapy pre irradiation protocol (70.6% vs 41.7%), which was used until the early 1990s when it was replaced by the COGA9661. Current strategies recommend risk-adapted CSI and adjuvant chemotherapy in children above 3-5 years(1), while infants should receive treatment with either high-dose chemotherapy and stem cell transplant(32) or intraventricular chemotherapy(33), in order to avoid or delay the use of radiation therapy.
The epidemiological characteristics of this series possibly explain the differences in survival that medulloblastoma patients have in Brazil. Issues related to limited health care resources, poverty, delayed diagnosis, treatment abandonment, and malnutrition are reflected in inferior survival outcomes when compared to high-income countries(34).There is a difference in survival between high-income countries treating their patients per COGA9661, with a reported 5-year OS of 87%(20) and the present study where the overall survival was 69.4% for SR patients. Despite the difficulties encountered in an upper-middle income country, it was possible to deliver treatment with good results. Nevertheless many initiatives can hopefully be explored to aid in improving the results and reducing these differences.
The limitations of the study are primarily due to its retrospective nature and the limitations of chart information, with missing data. Patients being operated in other institutions pose also difficulties, because of lack of information. The strengths of the study are the relatively large number of patients with medulloblastoma from a single institution, and the homogeneous treatment that most patients received. INCA as public reference hospital for cancer pediatric patients registers the vast majority of medulloblastoma patients in the state. Thus, this study is possibly a proxy of the medulloblastoma scenario in Rio de Janeiro.