Targeted Treatment of Soft-Tissue Sarcoma

Background: Soft-tissue sarcoma (STS) is a heterogeneous group of sarcomas with a low incidence. The treatment of advanced disease is poor, and mortality is high. We aimed to generate an overview of the clinical experiences with targeted treatments based on a pre-specified target in patients with STS. Methods: A systematic literature search was conducted in PubMed and Embase databases. The programs ENDNOTE and COVIDENCE were used for data management. The literature was screened to assess the article’s eligibility for inclusion. Results: Twenty-eight targeted agents were used to treat 80 patients with advanced STS and a known pre-specified genetic alteration. MDM2 inhibitors were the most-studied drug (n = 19), followed by crizotinib (n = 9), ceritinib (n = 8), and 90Y-OTSA (n = 8). All patients treated with the MDM2 inhibitor achieved a treatment response of stable disease (SD) or better with a treatment duration of 4 to 83 months. For the remaining drugs, a more mixed response was observed. The evidence is low because most studies were case reports or cohort studies, where only a few STS patients were included. Conclusions: Many targeted agents can precisely target specific genetic alterations in advanced STS. The MDM2 inhibitor has shown promising results.


Introduction
Sarcomas are neoplasms originating from the connective tissue. With more than eighty histological subtypes, sarcomas are highly heterogenous and rare, representing only about 1-2% of all adult cancers [1,2]. Sarcomas are divided into soft-tissue sarcoma (STS), representing approximately 84% of sarcoma patients, and sarcomas from bone and cartilage, representing approximately 16% of the patients [3]. The standard treatment of localised disease is surgery, with or without radiotherapy [4]. Despite a curative-intended therapy, the prognosis is grave as 25% of the patients will develop metastatic disease. These patients are primarily treated with chemotherapy [5,6]. However, the response to chemotherapy is low, and the mortality is high [2]. Therefore, the need for improved treatment for patients with advanced STS is needed.
In STS, the frequency of genetic alterations is between 84 and 91%, with the most frequently altered genes being TP53, ATRX, RB1, PTEN, MDM2, CDK4, and CDKN2A/B [2,8,11,13,[16][17][18][19]. Different genetic alterations can enable targeted therapy in patients suffering from advanced STS. However, while much is currently known about the genetic landscape of sarcoma, no overview covering treatment options against specific genetic alterations in STS exists. This systematic review aims to generate an overview of the clinical experiences with targeted treatments based on a pre-specified target in patients with STS. This study focuses on the outcome after targeted treatment of rare genetic alterations in sarcoma patients, not the genetic alteration by themselves or already effective treatments. Genetic testing is primarily performed in sarcoma patients with locally advanced or metastatic diseases treated with known standard palliative treatment. The studies included in this systematic review were selected for treatment given based on the results from comprehensive genetic testing.

Data Sources and Search Strings
A systematic review of the existing literature investigating the targeted treatment of adult STS was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines [20]. A comprehensive literature search in the medical databases PubMed and Embase was conducted on 1 April 2022. The following filters were used in both databases: "not animals" and "English". There was no time restriction. The search strings used are presented in Table 1.

Study Selection
The inclusion criteria were as follows: (I) original data, (II) the patients had to have a proven pathological STS with a specific genetic alteration prior to therapy with a targeted drug, (III) treatment outcome of one or several targeted agents had to be presented, (IV) if a study also included other types of cancer, the treatment outcome on STS had to be presented separately, (V) the patients had to be older than 15 years old, (VI) if a study included both paediatric and adult STS, the treatment outcome regarding adult STS had to be presented separately. Exclusion criteria were as follows: (I) non-English articles, (II) conference abstracts, (III) animal or in vitro studies, (IV) bone or cartilage sarcomas, hemopoietic sarcomas, and sarcomatoid tumours, (V) results only presented in figures, (VI) studies regarding immunotherapy. Patients with gastrointestinal stomal tumours (GISTs) treated with tyrosine kinase inhibitors (TKIs) were not included in the systemic review, even though they are treated with TKIs targeting different tyrosine kinases. The evidence for this treatment is well known, and for a systematic review of treating GIST, we referred to a systematic review published by Brinch et al. [21]. The multi-targeted drug pazopanib, a tyrosine kinase inhibitor targeting VEGFR, a platelet-derived growth factor receptor, and a c-kit were not included, as the treatment is given to many sarcoma patients with genetic testing with good clinical responses [22][23][24][25][26][27][28]. The same is true for regorafenib [29,30].

Data Extraction and Quality Assessment
All titles and abstracts were screened to identify eligible articles. One hundred studies were randomly selected and screened independently by all four authors to validate the abovementioned in-and exclusion criteria. Any disagreement was resolved by consensus.
Two of the authors (AIR and VNM) screened the rest of the titles and abstracts. In the case of disagreement, conflicts were resolved by all four authors.
After the identification of eligible full texts, ten randomly selected articles were read by three authors (AIR, BSP, and NAP) and subsequently included or excluded. The rest were included or excluded through full-text reading by one author (AIR) and, in the case of doubt, discussed by the other authors until a consensus was reached. Covidence (covidence.org) and Endnote (Clarivate Analytics) were used for duplicate and reference management during the inclusion and exclusion processes. The protocol was submitted to the PROSPERO database (CRD42021252341).
One author (AIR) performed data extraction, which subsequently was checked by the other authors. Data extraction included the first author's name, year of publishing, study design, genetic alteration that served as a target, type and/or name of the targeted drug, population, and treatment outcome. Studies included in this systematic review were quality scored by all authors using the Quality Assessment Tools for Case Series Studies and the Quality Assessment Tools for Observational Cohort and Cross-Sectional Studies, National Institute of Health, USA (https://www.nhlbi.nih.gov/health-topics/ study-quality-assessment-tools, accessed on 1 February 2022). The studies were rated "good", "fair", or "poor" according to the estimated risk of bias. All authors performed the quality assessment, and any disagreements were solved by consensus. Due to the heterogeneity of the studies and the many different outcomes they used, a meta-analysis could not be performed.

Study Selection
A total of 3996 titles and abstracts were identified using the two search strings presented in Table 1. After duplication screening, 493 duplicates were removed, resulting in 3503 unique hits. All titles and abstracts were screened, and subsequently, 169 full texts were read, returning 31 articles that met the abovementioned inclusion and exclusion criteria and, thus, were included in this systematic review. The inclusion and exclusion process is presented in Figure 1.
All titles and abstracts were screened to identify eligible articles. One hundred studies were randomly selected and screened independently by all four authors to validate the abovementioned in-and exclusion criteria. Any disagreement was resolved by consensus. Two of the authors (AIR and VNM) screened the rest of the titles and abstracts. In the case of disagreement, conflicts were resolved by all four authors.
After the identification of eligible full texts, ten randomly selected articles were read by three authors (AIR, BSP, and NAP) and subsequently included or excluded. The rest were included or excluded through full-text reading by one author (AIR) and, in the case of doubt, discussed by the other authors until a consensus was reached. Covidence (covidence.org) and Endnote (Clarivate Analytics) were used for duplicate and reference management during the inclusion and exclusion processes. The protocol was submitted to the PROSPERO database (CRD42021252341).
One author (AIR) performed data extraction, which subsequently was checked by the other authors. Data extraction included the first author's name, year of publishing, study design, genetic alteration that served as a target, type and/or name of the targeted drug, population, and treatment outcome. Studies included in this systematic review were quality scored by all authors using the Quality Assessment Tools for Case Series Studies and the Quality Assessment Tools for Observational Cohort and Cross-Sectional Studies, National Institute of Health, USA (https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools, accessed on 1 February 2022). The studies were rated "good", "fair", or "poor" according to the estimated risk of bias. All authors performed the quality assessment, and any disagreements were solved by consensus. Due to the heterogeneity of the studies and the many different outcomes they used, a meta-analysis could not be performed.

Study Selection
A total of 3996 titles and abstracts were identified using the two search strings presented in Table 1. After duplication screening, 493 duplicates were removed, resulting in 3503 unique hits. All titles and abstracts were screened, and subsequently, 169 full texts were read, returning 31 articles that met the abovementioned inclusion and exclusion criteria and, thus, were included in this systematic review. The inclusion and exclusion process is presented in Figure 1.

Study Description and Quality Assessment
Study characteristics for all included studies are presented in Table 2. All studies were either case reports (n = 17) or cohort studies (n = 14). In addition, an update of a case report was identified, and the sum of the two identified case reports was included [31,32]. The articles were published between 2015 and 2021. Twelve case reports were rated good, and the last five were rated as fair. Eleven cohort studies were rated good, two fair, and one poor.
In total, 80 patients representing 30 different histological subtypes were treated with targeted therapy based on a pre-specified target. Twenty-eight different drugs were used, targeting forty-one different targets. Fourteen patients received two or more different targeted agents in the same treatment course. Thirteen of the drugs were given only to one patient. The treatment effect on patients with advanced STS was measured by nine different treatment outcomes: complete response (CR), median time to progression, no evidence of disease (NED), overall survival (OS), partial response (PR), progression-free survival (PFS), progressive disease (PD), relapse-free survival, and stable disease (SD). In total, seven patients achieved CR, near CR, or NED. The patients achieving CR were treated as follows: Three patients received crizotinib targeting an NTRK1-KHDRBS1 fusion, ALK rearrangements, or an LMNA-NTRK1 fusion [32][33][34]; one patient received the MDM2 inhibitor, targeting an MDM2 amplification [11], and one patient received larotrectinib against a SPECC1L-NTRK3 fusion [35]. The two patients achieving NED received crizotinib against ALK expression and a SLC12A1-ROS1 fusion, respectively [36]. Only the MDM2 inhibitor, crizotinib, ceritinib, and 90 Y-OTSA were given to eight or more patients.

MDM2 Inhibitors
MDM2 inhibitors were given to 19 patients. All patients receiving MDM2 inhibitors were patients with liposarcoma and MDM2 amplification; furthermore, all patients had a treatment effect of SD or better with a treatment duration ranging from 4 to 83 months [11,13,37].

90 Y-OTSA
Eight patients with FZD10 expression received the targeted radioactive drug 90 Y-OTSA. Three patients received 370 MBq of these; one patient had SD, and two patients PD. Five patients received 1100 MBq; two had SD, and three had PD. One of the patients with SD receiving 1100 MB1 90 Y-OTSA also received a second injection that resulted in PFS for 21.4 weeks [45].

Discussion
Advanced sarcoma is a rare but severe disease, and better treatment options are needed. We have conducted a systematic review to cover targeted treatment for patients suffering from advanced STS with a known genetic alteration.
The use of targeted treatment in STS is still on an individual and experimental level. Most of the studies included in this systematic review were case reports or cohort studies where only a few individuals suffering from advanced STS were included. Summarised, 80 patients received 28 different targeted drugs. Thirty different subtypes of STS received targeted treatment, and among them, forty-one different genetic alterations were identified. Thus, several targeted agents were administered to patients with different kinds of advanced STS but with highly variable results.
Several studies have shown that there are many different genetic alterations in STS. Groisberg et al. found that 93 out of 102 advanced STS patients had at least one genetic alteration, where the most frequent were found in TP53, CDK4, and MDM2, and 61% had a potentially targetable alteration. However, only 16% of the patients received a targeted drug [11]. Similar observations were found by Lucchesi et al., where 84% of the 584 included STS patients enharboured a genetic alteration. Furthermore, the most frequent mutations were observed in TP53, MDM2, and CDK4, and 41% of the patients had at least one potentially targetable alteration [18]. Gusho et al. also found a high frequency of genetic alterations with at least 1 genetic alteration in 122 out of 136 samples. The most frequent mutations observed were in TP53, CDKN2A/B, and RB1, and 47% had a potentially targetable mutation [19].
In this review, all patients receiving MDM2 inhibitors had liposarcomas, with a treatment response of SD or better, including one CR [11,13,37]. MDM2 amplification is a common genetic alteration in STS. Three studies have shown a 5%, 22%, and 40% prevalence of STS [2,11,13]. In specific subtypes of STS, the prevalence is even higher. In well-differentiated and dedifferentiated liposarcoma, the amplification of 12q13-15 is common, resulting in the amplification of the MDM2 gene. Studies have shown that over 90% of these sarcomas have MDM2 amplification [13,37]. Because of the high prevalence of MDM2 alterations in STS, especially in liposarcoma, and the reports with high clinical effects, targeting this specific alteration must be considered.
Crizotinib was given to patients with mutations in ALK, ROS1, MET, or NTRK1. There were varying treatment responses, but three out of nine patients had CR or near-CR, and two patients had NED; therefore, they had an overall survival benefit. One of the patients with CR had ALK gene rearrangements, and the other had NTRK1-KHDRBS1 fusion. The patient with a near-CR response had an LMNA-NTRK1 fusion. The patients with NED had high ALK expression and a SLC12A1-ROS1 fusion [32][33][34]36].
Owing to the aim of this study, only patients receiving treatments targeting a known specific target were included. Therefore, only seven of the included patients received pazopanib, a multi-target tyrosine kinase inhibitor; however, the drug is given to far more patients suffering from advanced STS. Even though studies exist on pazopanib in STS, these studies were not included as they did not have a specific pathological proven target before treatment with pazopanib. Many of these studies showed a treatment effect on STS without prior identification of a specific genetic alteration.
MDM2 inhibitors, crizotinib, ceritinib, and 90 Y-OTSA were given to most patients. However, the population receiving targeted therapy is so tiny that it is difficult to conclude the overall survival effect of the targeted agents. Furthermore, nine different treatment outcomes were used to measure the effectiveness of the drugs, making it challenging to conduct a meta-analysis to investigate the potential survival effect of agents on advanced STS.
Most of the included patients were previously treated with curative-intended treatments, and many had one or more treatment courses with chemotherapy. Furthermore, targeted agents were tried as a last resort for patients with advanced STS. Maybe, if the targeted therapy had been introduced to the patients before relapse occurred, the patients might have had a more substantial benefit from the treatment and more prolonged survival. It would have been interesting to compare targeted treatment against chemotherapy as the first-line treatment for advanced STS to see if there is a difference. Larger studies are needed to investigate the potential effect of targeted therapy on STS, for example, case/control studies, where targeted treatment is compared to chemotherapy and ultimately randomised controlled trials.
The strengths of this systematic review are the comprehensive systematic review of the available literature on targeted treatment of STS in two major medical databases. Broad search strings and in-and exclusion criteria were used to avoid missing relevant articles. Two authors screened the literature, and all authors validated the in-and exclusion criteria. Nonetheless, this review also has its limitations. All the studies included are either case reports or cohort studies, where only a few patients have advanced STS. The studies used many different measures of outcome. One of the exclusion criteria was non-English articles, and it is possible that other studies concerning the same topic were written in other languages. Thus, language bias cannot be excluded.

Conclusions
During the last decades, many genetic alterations in STS have been identified, enabling the use of targeted treatment. This systematic review revealed many studies regarding genetic alterations and targeted treatment in advanced STS. Twenty-eight targeted agents have currently been tried in advanced STS. However, most articles are case reports and cohort studies representing only a few patients with advanced STS. Studies comparing targeted treatment to chemotherapy in case/control studies and ultimately randomised controlled trials can make it easier to investigate the potential survival benefit that targeted agents can provide to patients suffering from advanced STS.

Conflicts of Interest:
The authors declare no conflict of interest.