Survival Benefit from Multimodal Treatment for Patients with Atypical Teratoid Rhabdoid Tumor in a Surveillance, Epidemiology, and End Results Database Analysis Free

Atypical teratoid rhabdoid tumor (ATRT) is defined as a highly malignant embryonal neoplasm of the central nervous system (CNS) [1]. The World Health Organization classifies this as a grade 4 malignancy [2]. Typically, this tumor affects children aged less than 3 years; however, there have been a handful of rare cases reported in adults [3, 4].

In the past, ATRT was commonly misreported as medulloblastoma or a primitive neuroectodermal tumor. This is mostly due to the fact that ATRT has a complex histologic pattern of a combination of rhabdoid, epithelial, primitive neuroepithelial, and mesenchymal components. These histological patterns lead it to be commonly mistaken for medulloblastoma, choroid plexus carcinoma, or CNS primitive neuroectodermal tumor [5]. The ability to diagnose ATRT has only recently been made accessible due to the advent of immunohistochemical testing in the late 1990s [6]. Loss of expression of the INI-1 protein is sensitive and specific and has become a reliable method for diagnosing ATRT tumors [7].

ATRT reportedly has a 5% incidence among CNS tumors in children aged greater than 18 years and up to 20% incidence among CNS tumors in children aged less than 3 years [8, 9]. The true incidence might even be higher due to the instances when ATRT has been misdiagnosed as another high-grade embryonal CNS malignancy [10].

There is limited literature on the clinical characteristics, treatment options, and prognosis for ATRT. Additionally, the prognosis for ATRT remains dismal. Historically, patients with ATRT have an overall survival (OS) that ranges from 6 to 18 months [8]. Targeted therapies have been based primarily on the treatment for other neuroendocrine tumors and medulloblastomas. However, the challenge has been that these tumors have a considerably better prognosis, and therapeutic regimens have not been translatable to ATRT [11]. Currently, there is no standard therapeutic regimen for ATRT [12].

The National Cancer Institute (NCI) Surveillance, Epidemiology, and End Results (SEER) database collects critical data on incidence, survival, outcomes, and demographics from population-based cancer registries. Given that their database includes approximately 28% of the USA population, it provides the unique opportunity to study the clinical characteristics, treatment regimens, and outcomes on a larger scale for rare entities, such as ATRT [13]. The objective of the current study was to classify the prognostic features of ATRT and classify the benefits of utilizing certain therapies. We also hope to highlight some of the advents that have been made available over the past decade to treat this rare malignancy.

This study examines data from the NCI SEER 17 collected from 2000 to 2019. Patients were identified using the International Classification of Diseases for Oncology, version 3 (ICD-O-3) morphology code 9508/3. All data used in this dataset are publicly available. Therefore, this study was deemed exempt from Institutional Review Board at Carilion Clinic oversight and approval as the referenced project did not meet the definition of human subject research (IRB-23-1867). This study complies with the NCI SEER limited-use data end-user agreement.

Patient sex, age at diagnosis, race, year of diagnosis, primary tumor site, tumor size, time in months from diagnosis to treatment, and treatment type were the clinical characteristics that were evaluated. Patients received varying combinations of treatment types. Surgery, radiation treatment, and chemotherapy were also examined as separate categories to see how specific treatments individually impacted patient outcomes. Our database also provided information on the extent of surgical resection which was included in our analysis. Our database did not provide sufficient data on the type of radiation therapy and type of chemotherapy agents used. This is addressed later in the manuscript as a limitation. A patient’s age at diagnosis was only available in years; thus, an age of 0 was defined as the patient’s age between birth and 1 day before age 1. The primary endpoint used for this study was OS defined as the time from diagnosis until the time of death, as reported by SEER.

Each of the categories was assessed using Cox proportional hazard analysis using the Breslow method for ties. First, univariate survival analysis was performed on all categories. Any categories that showed statistical significance (p value <0.05) were included in subsequent multivariable survival analyses. Two multivariable analyses were performed to analyze the effects of individual therapies and combination therapies on survival. The Kaplan-Meier method was employed to visualize a subset of survival curves corresponding to the categories that showed significant differences in OS based on the log-rank test. All statistical analyses were performed on RStudio (version 4.2.2) using the “survival” and “survminer” packages.

In total, the retrospective cohort included 363 patients. Three patients were excluded due to insufficient information about their OS. Our cohort included slightly more males (n = 191; 53%). The median age at diagnosis was 1 year (range: 0–74 years). The majority of patients were ≤ 3 years old (78%). Fourteen individuals were adults (4%), ranging in age between 19 years old and 74 years old. Most of the patients were non-Hispanic white (50%). African American, Asian, and Hispanic represented 12%, 10%, and 27%, respectively.

The reported primary tumor sites were well dispersed throughout the CNS (Table 1). Besides 1 patient, all patients could be grouped into two major categories of the primary tumor site: supratentorial and infratentorial. Most patients had a primary tumor site in the supratentorial region (n = 223; 61%). The tumor size at diagnosis was available for 74% of the patients. Among those patients (n = 269), 21% of the tumors were ≤3.0 cm (n = 56), 56% of tumors were between 3.1 cm and 6.0 cm (n = 151), and 23% of the tumors were ≥6.0 cm (n = 62) by the measurement of the largest diameter length.

Among the cohort, the three primary reported therapies that were used to treat patients with ATRT include surgery, beam radiation therapy, and chemotherapy. Twenty-two patients received no treatment at all (6%). Only 61 patients received surgery without receiving any chemotherapy or radiation (17%). Nine patients received only chemotherapy for treatment (2%). No patients received only radiation for treatment. The majority of patients in our cohort received a combination of treatments. Most of the patients received surgery, radiation, and chemotherapy (n = 138; 38%). Thirty-six percent of the patients received surgery and chemotherapy (n = 119). Ten patients underwent surgery and radiation treatment (3%). Four of the patients received radiation and chemotherapy (1%). Overall, 90% of the patients received surgical treatment (n = 328). The 10% that did not receive surgical treatment had a biopsy to confirm the diagnosis. Seventy-four percent of patients received chemotherapy (n = 270). Forty-two percent of patients received beam radiation therapy (n = 152). A summary of patient demographics, tumor characteristics, and treatment regimens is depicted in Table 1.

Through the use of Cox proportional hazards modeling, clinical factors that had a large impact on OS with the univariate analysis were identified (Table 2). The univariate analyses by category revealed that the age at diagnosis, year of diagnosis, receipt of surgical treatment, receipt of beam radiation, receipt of chemotherapy, and receipt of certain combinations of therapies all had a statistically significant (p < 0.05) impact on OS. Ages 1–3 and 4–17 were correlated with a better prognosis. Patients who were diagnosed after 2010 had a better prognosis for survival compared to those who were diagnosed before 2010. Patients who received surgical treatment, beam radiation, or chemotherapy had a better OS. Patients who received any type of combination of the three treatments were less likely to die compared to those who received no treatment at all or those who received only one form of treatment. Of note, there were no patients who received only beam radiation therapy.

Two different multivariable analyses were conducted. The first multivariable analysis considered the age at diagnosis, year of diagnosis, receipt of gross total resection versus subtotal resection, receipt of radiation therapy, and receipt of chemotherapy. The year of diagnosis no longer showed any significance. There was a borderline significance (p = 0.053) for patients aged 4–17 years. The remaining groups still held significance in their impact on survival (Fig. 1a). The second multivariable analysis considered similar prognostic factors of the age at diagnosis and year of diagnosis but instead included the receipt of different combined therapies. Similarly to the prior multivariable analysis, this showed that the year of diagnosis no longer had significance on survival. Additionally, there was no longer significance on survival for the age of diagnosis between 4 and 17 years. Meanwhile, there was still significance seen in the age of diagnosis between 1 and 3 years. Each of the combined therapy groups maintained a strong impact on OS (Fig. 1b).

This retrospective cohort had a median OS of 13 months (95% confidence interval [CI], 10–19 months) based on the Kaplan-Meier method (Fig. 2a). Patients who were between the ages of 1 and 3 years and 4 and 17 years old at diagnosis had better median OS (18 months with 95% CI: 11–59 months, and 37 months with 95% CI: 23 months to not yet reached, respectively; log-rank p < 0.001) compared to the median OS of patients aged less than 1 year (6 months; 95% CI: 5–9 months; log-rank p < 0.001) (Fig. 2b). Although being diagnosed in some of the more recent years showed a slightly significant impact on the survival of patients, this did not hold true in either of the multivariable analyses. For this reason, we concluded that the year of diagnosis did not reach a statistically significant effect on survival (data not shown). Patient sex, race, tumor site, and tumor size also did not reach a statistically significant effect on survival (data not shown).

Those patients who received surgical therapy had a higher median OS than those who did not receive any surgical therapy (17 months with 95% CI: 13–25 months vs. 1 month with 95% CI: 1–4 months; log-rank p < 0.001). This was further analyzed by the extent of surgery. Those patients who received gross total resection and subtotal resection had significantly higher median OS than those who did not receive any surgical therapy (23 months with 95% CI: 15–69 months and 13 months with 95% CI: 10–22 months, respectively; log-rank p < 0.001) (Fig. 3a). For those patients who did not receive any radiation therapy, the median OS was 5 months (95% CI, 4–7 months; log-rank p < 0.001), whereas the median OS for patients who did receive radiation therapy was not yet reached at the time patients were reported (95% CI, from 52 weeks to not yet reached; log-rank p < 0.001) (Fig. 3b). Those patients who received chemotherapy had a much higher median OS than those who did not receive any chemotherapy (24 months with 95% CI: 18–43 months vs. 2 months with 95% CI: 1–3 months; log-rank p < 0.001) (Fig. 3c).

Another subsequent analysis was conducted to analyze the effect of different combinations of therapies on OS. The median OS significantly improved with any of the different combinations of therapies. Patients who received surgical therapy and chemotherapy had a median OS of 10 months (95% CI, 8–14 months; log-rank p < 0.001). Patients who received surgical therapy and radiation therapy had a median OS of 35 months (95% CI, 4 months to not yet reached; log-rank p < 0.001). Patients who received radiation therapy and chemotherapy had a median OS of 13 months (95% CI, 4 months to not yet reached; log-rank p < 0.001). Finally, patients who received surgical therapy, radiation therapy, and chemotherapy had a median OS that was not yet reached at the time patients were reported (95% CI, 55 months to not yet reached; log-rank p < 0.001). Those patients who received no treatment had an OS that was significantly lower than that of any of the combination therapy groups (1 month; 95% CI, 1–2 months; log-rank p < 0.001). The median OS for patients who received only surgical therapy was also significantly lower than that of any of the combination therapy groups (2 months; 95% CI, 1–3 months; log-rank p < 0.001). Those patients who received only chemotherapy also had a lower median OS than any of the combination therapy groups (1 month; 95% CI, 1 month to not yet reached; log-rank p < 0.001). The survival curves of each of these different combinations of therapies are shown in Figure 4.

ATRT is a rare malignant neoplasm that primarily presents at an early age. Currently, there is no standardized treatment protocol for patients with ATRT [12]. Surgical resection has been the main modality of treatment for these patients for many years [14‒16]. Multiple studies have shown that patients who undergo gross total resection of their tumor have the best survival benefits [14, 17]. However, depending on the location of the tumor, size of the tumor, and overall disease burden, achieving a radical resection may not be possible [14]. Although the extent of surgical resection improves OS for patients as demonstrated in our study, surgical resection alone may not be enough. Adjuvant therapy can provide further improvement of OS.

In recent years, the utilization of radiation and chemotherapy has increased as a treatment for ATRT, which has been shown to improve OS [18]. Based on previous studies, the current chemotherapies that are used to target ATRT include high-dose chemotherapy with peripheral blood stem cell rescue and methotrexate [19]. In addition, alkylating and anthracycline agents that have traditionally been effective against rhabdomyosarcoma have been extrapolated to treat ATRT as well [8]. Studies investigating the role of adjuvant chemotherapy and radiation in ATRT have shown promising results. Intensive chemotherapy regimens, such as the Children’s Cancer Group (CCG)-9921 and Intergroup Rhabdomyosarcoma III (IRS III) protocols, have been used, leading to overall response rates of around 42% [20]. The CCG-9921 induction chemotherapy regimen includes either vincristine, cisplatin, cyclophosphamide, and etoposide or vincristine, carboplatin, ifosfamide, and etoposide [20]. The IRS III protocol includes vincristine, cisplatin, doxorubicin, cyclophosphamide, dacarbazine, etoposide, actinomycin-D, and triple intrathecal chemotherapy with methotrexate, hydrocortisone, and cytarabine [21, 22]. Some studies explored the possibility of deferring radiation therapy in young children by using high-dose methotrexate-based multidrug chemotherapy and autologous stem cell rescue, resulting in long-term survival in a subset of patients without irradiation [23, 24]. However, the use of high-dose chemotherapy with stem cell rescue should be carefully considered due to its formidable toxicity profile [24].

While adjuvant chemotherapy has shown significant efficacy in ATRT treatment, the disease is highly aggressive, with frequent recurrence within 6 months. To address recurrence, some studies have used single-agent temozolomide, leading to prolonged disease stabilization without severe toxicity [25, 26]. Despite these advancements in chemotherapy agents, ATRT remains a challenging malignancy, and further research is needed to optimize treatment strategies and improve long-term survival rates for affected patients.

Given that the majority of patients diagnosed with ATRT are younger than 3 years old, the effect of radiation therapy on neurocognitive development has deterred providers from utilizing this method of treatment in very young patients [12]. Recent studies have shown benefits to patients when radiation therapy was started earlier in their disease course [8]. The required radiotherapy dose for ATRT patients is not standardized due to the tumor’s rarity and limited use of radiotherapy in many cases. Typically, the dose to the tumor bed ranges from 50 Gy to 56 Gy, while the dose to the neuraxis ranges from 23.4 Gy to 36 Gy in conventional fractionation [27, 28]. In retrospective series studies, postoperative radiotherapy was started and completed in some patients, with the use of craniospinal irradiation being a significant predictor of OS in one study [26]. Another study found that a total radiotherapy dose of >50 Gy was associated with significantly improved failure-free survival but not OS on multivariate analysis [28].

Research is also being conducted to investigate the benefits of molecular-based targeted therapies and immunomodulator therapies [29‒31]. In a preclinical study conducted by Jayanthan et al. [32], they explored the potential of multikinase inhibitors in treating ATRT. The researchers identified the expression of specific receptor tyrosine kinases (c-Kit, platelet-derived growth factor receptor β, vascular endothelial growth factor receptor 2, and Flt3) in ATRT cell lines. Multikinase inhibitors, namely, sorafenib and sunitinib, were found to effectively inhibit these target receptors. The study demonstrated that these inhibitors could dose-dependently impede the growth of ATRT cells. Additionally, there was evidence of synergistic effects when combining irinotecan with sorafenib and sunitinib, possibly due to irinotecan’s ability to enhance vascular endothelial growth factor-directed therapy.

Another study by Sredni et al. [33] analyzed gene expression profiling in ATRT and found the overexpression of ErbB2 and ErbB3, leading to the downstream activation of the Ras/Raf/MEK/ERK pathway. To target these receptors, the researchers used lapatinib, a dual tyrosine kinase inhibitor of ErbB1 and ErbB2. The study observed progressive cell death in ATRT cell lines with increasing concentrations and exposure time to lapatinib. Furthermore, Fouladi et al. [34] conducted a phase II trial to evaluate the effectiveness of lapatinib in treating children with refractory CNS malignancies, including recurrent medulloblastoma, ependymoma, and high-grade glioma. Although lapatinib was well tolerated in these children, it showed limited activity when used as a single-agent therapy. In summary, these preclinical and clinical studies shed light on the potential role of multikinase inhibitors, especially lapatinib, in treating ATRT, suggesting promising avenues for further investigation and therapeutic approaches for this challenging malignancy.

Of our cohort of 363 patients with ATRT, 328 (90%) received surgical treatment. Furthermore, only 61 patients received surgery only. These patients did not have a significant improvement in their OS compared to the patients who received no treatment. As mentioned above, the median survival for patients who received surgery was only 2 months. Although most of these patients who only received surgical treatment fall under 3 years of age, the multivariable analysis did not show any significance in mortality. Bachu et al. [18] showed similar findings that sole surgical treatment had a median survival of about 6 months among a retrospective cohort of 189 patients. This further proves the importance of adjuvant therapy in the treatment of ATRT. Based on our data, it seems that radiation therapy alone is not seen as a primary mode of treatment. In our cohort, 270 (74%) patients received chemotherapy, among which only 9 patients were those who received chemotherapy as their only method of treatment. These patients did not have a statistically significant improvement in OS when compared to those who did not receive any treatment on multivariable. It is unlikely that patients receive only chemotherapy; however, given that they were of a younger age (mean = 5 years), they likely did not qualify for surgical or radiation therapy.

In our cohort, 271 (75%) patients received some form of combination therapy: surgery and chemotherapy, surgery and radiation, chemotherapy and radiation, and triple therapy. As shown above, each of these combination therapies had a statistically significant improvement in OS when compared to those who received no treatment on both univariate and multivariable analyses. Among our cohort, 138 (38%) patients received triple therapy. The median OS was almost 2-fold for those who received triple therapy (55 months) when compared to the combination of other therapies. Bachu et al. [18] demonstrated that triple therapy had a median survival of 68 months; however, their cohort only included 16 patients who received triple therapy. Given that there is no standard of care therapy plan for the treatment of ATRT, based on our results, triple therapy should be considered for patients who meet the criteria.

The overall treatment for ATRT has been adapting over time. Our hypothesis was that the OS of patients would change based on their year of diagnosis, assuming that our cohort is receiving the current and most advanced treatment. Although this trend was proven to be statistically significant in our univariate analysis, when it was factored into our multivariable analysis, it lost significance. Based on our data, there needs to be further investigation on advancing the treatment for ATRT because, currently, the OS rate has not been significantly different between the years 2000 and 2018. Similarly, we expected there to be a significant difference in survival for the location of the tumor and size of the tumor. However, this was not the case in univariate analysis. We believe this is probably due to the aggressive nature of ATRT [31, 35]. In addition, different genetic studies have shown that the supratentorial and infratentorial tumors have varying genetic profiles that might contribute to OS [35].

The SEER database is an excellent resource for studying rare tumors, such as ATRT; however, this approach has its own limitations. Our database does not go into detail on the type of radiation and chemotherapy received, which could be crucial in influencing the OS rate. Other data on the progression of disease post-treatment, perioperative performance status, recurrence rate of tumor, and repeat surgery were not included in this database. Another limitation that is inherent to this analysis is the fact that there is no information on the timing of treatment. Even though the database reports the time to treatment from diagnosis, it is not possible to ascertain which treatment was started first or the timing between each treatment, specifically for the participants who receive combination therapy. All these factors would provide a better understanding of clinical features and overall performance of the therapies.

This paper is a comprehensive retrospective analysis of patients with ATRT. We concluded that patients who received any kind of combination therapy had an improvement in their OS. Despite the OS being generally low for this patient population, the most important finding of our study is that patients who received triple therapy had significantly better OS. These data will shed light on some of the key attributes of ATRT and guide future treatment strategies.

An ethics statement is not applicable for this study because all data used in this dataset are publicly available. This study was deemed exempt from Institutional Review Board at Carilion Clinic oversight and approval as the referenced project did not meet the definition of human subject research (IRB-23-1867). Additionally, for the same reason, written informed consent was not required for this study. This study complies with the NCI SEER limited-use data end-user agreement.

The authors have no conflicts of interest to declare.

No funding was received in any form for this work.

All of the authors provided substantial contributions to the conception or design of the work or the acquisition, analysis, or interpretation of data for the work; drafted the work or revised it critically for important intellectual content; provided final approval of the version to be published; and have agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Conceptualization, investigation, validation, visualization, and writing – original draft and review and edition: A.S.B., S.A., J.A.C., C.M.R., and E.A.M. Data curation: A.S.B. and S.A. Formal analysis and methodology: A.S.B. S.A., and J.A.C. Supervision: J.A.C., C.M.R., and E.A.M.

Abhishek S. Bhutada and Srijan Adhikari contributed equally to this work and share first-authorship.

No new data were generated or analyzed in support of this research. All data used for this analysis are publicly available through the NCI SEER database. All data generated during this study are included in this article. Further inquiries can be directed to the corresponding author.