The study of MTTs, like any rare tumour type, has largely been limited to case reports in order to guide care. The highest level of evidence possible is a systematic review such as the one we have carried out. Prior to our report, a study by McConnell et al. in 2012, combined 99 adult and 25 paediatric MTT cases (1). With 124 adult cases, the current study provides a rigorous analysis of the updated literature. In addition to including a larger number of adult MTT cases, we restaged cases using the AJCC 8th edition staging system (which was feasible in all cases based on case descriptions) and examined the role of prior radiation exposure. Similarly, we found that margin status was associated with poorer PFS and adjuvant radiation with improved OS. In addition, we identified that positive NF1 status, prior radiation exposure and tumour location in the CNS or spine were also associated with poorer OS. In contrast to McConnell et al., tumour and nodal stage as well as increasing tumour size were not associated with survival on univariable or multivariable analysis(1). No survival benefit was observed for adjuvant chemotherapy.
NF1 and prior radiation exposure are well documented as risk factors for the development of MTT (1,5). Out of the 124 cases in this review, 103 case reports commented on NF1 status, and 35.0% of these were positive for NF1. This is consistent with Li et al. who found 41.7% of their cases had NF1(4). McConnell et al. found NF1 status to be a negative prognostic factor with increased mortality on univariable analysis, however it did not remain significant on multivariable analysis(1). In this review, NF1 status was a significant predictor of both OS and PFS on univariable and multivariable analysis. LaFemina et al. examined 49 sporadic, 42 NF1-associated, and 14 radiation-induced MPNSTs(6). Interestingly, they found that sporadic MPNSTs and NF1-associated MPNSTs had similar survival and recurrence outcomes(6). This is not in keeping with our findings; however, MPNST subtypes were not reported in the analysis by LaFemina et al.(6) Given our findings, it is possible that NF1 associated MTTs behave more aggressively than other NF1 associated MPNSTs. In the current analysis, 5 cases of previous radiation exposure were identified, and it was found to be a significant predictor of worse OS on both univariable and multivariable analysis, which was consistent with LaFemina et al. reporting worse oncological outcomes for patients with RT-induced MPNSTs(6).
The mean age of MTT diagnosis in this review was 42.8 years. Previous literature states that NF1 positive patients tend to present as young males, and sporadic cases of MTT present more frequently in older females(2,5). The mean age of male cases in our study population was 41.4 years, compared to 44.6 years for female cases. The majority of cases with a history of NF1 were male (n = 26, 72.2%), with a mean age of 35.2 years, whereas females with NF1 (n = 10, 27.8%) had a slightly older mean age of 37.8 years. A study conducted by Terzic et al., found a mean age of 26.2 years (range: 3–44) for their NF1 positive cases(5). This younger age of diagnosis for NF1 positive patients in the literature compared to the current study is largely attributed to the exclusion of paediatric cases.
Our systematic review provides a basis for treatment recommendations based on the management strategies and treatment outcomes observed. Almost all identified cases utilized surgery (95.9%) suggesting that this modality is a cornerstone of treatment. Comparable to most solid malignancies, positive surgical margins were a poor prognostic factor highlighting the importance of planning for and obtaining an R0 resection as was obtained in our case report. In this review, the most common tumour sites were head and neck and extremities. This is in keeping with other studies which found these tumour sites were common(2,4). CNS and spine tumour sites were significantly predictive of worse OS. This is likely due in part to difficulty resecting tumours in these sites as well as critical nearby organs limiting the potential adjuvant radiation dose.
Our study reinforces the critical importance of adjuvant radiotherapy. Radiotherapy was utilized in 60.7% of cases and was associated in significantly improved OS. Adjuvant radiotherapy was significant on univariable analysis for PFS, but it did not remain significant on multivariable analysis. Radiation dose varied, but most cases received doses greater than an equivalent dose in 2 Gy per fraction (EQD2) of 50 Gy. In contrast, chemotherapy was not found to be of benefit and was even associated with worse PFS on univariable analysis. These findings are likely confounded by the heterogeneity of chemotherapy drugs and dosing schedules used, as well as a bias towards using chemotherapy in more advanced cases with adverse pathological features. Likely more active agents are needed to improve the poor survival observed in our cohort.
Although recent studies suggest that immunotherapy may be effective in slowing progression of plexiform neurofibromas in NF(7), the role of immunotherapy in MPNSTs is unknown. There are case reports indicating that immunotherapy may provide some benefit(8); however, we do not yet have any high-level evidence to support its use. Fortunately, there are currently several clinical trials underway examining this question. An active phase 2 trial examining the effects of pembrolizumab (anti-PD-L1 antibody) in MPNST patients that do not qualify for curative surgery is currently recruiting with results expected in 2022 (NCT02691026)(9). A second phase 2 trial is examining the effectiveness of nivolumab (anti-PD-L1 antibody) and ipilimumab (anti-CTLA-4 antibody) in treating a wide variety of rare tumours and has closed to accrual for many of the tumour types including MPNST, with results expected in 2021 (NCT02834013)(10). A third study of 20 patients with unresectable or refractory MPNST’s (NCT02008877), treated with a combination of ganetespib and sirolimus was well tolerated but ineffective and did not meet criteria for further study(11).
Interestingly, T stage was not a significant predictor for either OS or PFS. This could be partially attributed to over 70% of our sample size presented at stage T3 or T4, or that tumour size may not be a factor that impacts the likelihood of local, regional or distant spread. In our review, only 4 cases had nodal disease and 1 patient had metastatic disease on presentation. N stage was also not a significant predictor for either OS or PFS, most likely due to the limited number of cases presenting with nodal involvement. This is in keeping with the literature that highlights presentations of MTT with nodal involvement or metastasis are rare(5,12). Divergent differentiation was present in 24 (19.8%) of the 121 cases in which it was reported. The most common types were chondroid (n = 5, 20.8%), osteocartilaginous (n = 5, 20.8%) and glandular (n = 4, 16.7%). Divergent differentiation was a significant predictor of worse OS on univariable analysis but did not remain significant on multivariable analysis.
Median and 5-year PFS was 8 months and 21.3%, respectively. Just over half of the cases in this study experienced recurrence of the neoplasm (n = 63, 50.8%), with the most common recurrence patterns being local only (n = 24, 38.1%), distant only (n = 23, 36.5%) and combined local and distant recurrences (n = 12, 19.1%). This is similar to McConnell et al., where 50% of their cases had a recurrence and the median PFS was 6 months(1).
Systematic reviews are inherently limited by the data which is obtainable only through literature review. The majority of the studies used for this review were case reports, allowing for more detailed data collection; however, there were still missing data for many of the cases. Further, there was a high risk of publication bias across the studies included in this review due to their nature as case reports, as there is likely an over-representation of complex cases or impressive treatment responses, as these were more likely to have been written as case reports. Despite these limitations, for rare tumours such as MTTs where randomized clinical studies are not feasible, systematic reviews provide clinicians with evidence-based prognostic data and treatment recommendations for their patients.