Reference
Response grade
Response grade definition
I
Viable; little to no chemotherapeutic effect
II
Partially necrotic
III
Largely necrotic
IV
Totally necrotic; no histologic evidence of viable tumor within specimen
Rosen et al. (1979)
I
Little to no chemotherapeutic effect
II
Partial response, <50 % necrosis, some viable tumor remaining
III
>90 % tumor necrosis, foci of viable tumor remaining
IV
No viable-appearing tumor cells
Raymond et al. (1987)
Percentage of necrosis estimated
Picci et al. (1994)
Good
>90 % necrosis
Fair
60-90 % necrosis
Poor
<60 % necrosis
Wold (1998)**
**Used in COG clinical trials
I
No chemotherapeutic effect
II
Some necrosis
A
50 % viable tumor remaining
B
5–50 % viable tumor remaining
III
Scattered foci; <5 % viable tumor remaining
IV
No viable tumor remaining
Salzer-Kuntschik et al. (1983)
I
No viable-appearing tumor cells
II
Single viable tumor cells or 1 viable cell cluster, <0.5 cm
III
<10 % Viable tumor remaining
IV
10–50 % Viable tumor remaining
V
>50 % Viable tumor remaining
VI
No chemotherapeutic effect
Table 8.2
Histologic response grading systems applied to treated Ewing sarcomas/PNETs
Reference | Response grade | Response grade definition |
|---|---|---|
Huvos (1988) | I | Viable; little to no chemotherapy effect |
II | Partially necrotic | |
III | Largely necrotic | |
IV | Totally necrotic; no histologic evidence of viable tumor within specimen | |
Salzer-Kuntschik et al. (1983) | I | No viable appearing tumor cells |
II | Single vital tumor cells or 1 vital cell cluster, <0.5 cm | |
III | Vital tumor, <10 % | |
IV | Vital tumor, 10–50 % | |
V | Vital tumor, >50 % | |
VI | No effect of chemotherapy | |
I | 1+ macroscopic nodule of viable tumor (>one 10× field) or Scattered microscopic nodules in summation (>one 10× field) | |
II | Isolated microscopic nodules of viable tumor cells in summation (<one 10× field) | |
III | No viable tumor cell nodules; scattered individual tumor cells permitted |
8.2.2 Prognostic Indications of Histological Response in Osteosarcoma
The phenomenon of prognostic histological response has been observed for osteosarcoma in pediatric patients as well as the adolescent and young adult (AYA) population alike. It has been recognized in several studies that pediatric patients with osteosarcoma who have a good histological response have significantly higher survival rates (5-year overall survival (OS) rate of approximately 76 %, 5-year event-free survival (EFS) rate of approximately 73 %), as compared to patients with less than 90 % postchemotherapy tumor necrosis, who have an OS of 48 % and EFS of 44 % (Bacci et al. 2001, 2002a, 2006b; Ferrari et al. 2001; Goorin et al. 2003; Halperin 2011). Utilizing cutoffs for postsurgery necrosis of 95 % or greater in those 18 years or older, a good response has been correlated to increased overall survival in the AYA population (Janeway et al. 2012).
With hopes to use histological response in order to determine an appropriate subsequent treatment course for patients with bone tumors, several clinical trials incorporated the determination of histological response and consequent patient risk stratification with therapeutic modulation into their protocols. However, a majority of the clinical trials failed to show significant improvement in patient outcomes using this as a risk stratification method. For instance, in the three neoadjuvant chemotherapeutic osteosarcoma studies performed at the MD Anderson Cancer Center, Treatment and Investigation of Osteosarcoma (TIOS) I-III, the preoperative chemotherapeutic response was used to plan postoperative treatment. The postchemotherapy percentage of tumor necrosis (using cutoff of 90 % or greater) was found to significantly correlate with relapse and the development of pulmonary metastases (p = 0.01) (Hudson et al. 1990). As another example, one of the principal objectives of the study CCG-782 was to use histological response of the primary tumor after neoadjuvant chemotherapy to determine the postoperative chemotherapy regimen. This study defined a cutoff of less than 95 % tumor necrosis as poor responders who demonstrated significantly higher risk for an adverse event as compared to good responders (relative risk 0.23, p < 0.0001) (Miser et al. 1993). In the POG 8561 trial, patients who had less than 10 % viable tumor after induction with chemotherapy were shown to have significantly improved EFS (73 %) when compared with patients with poor response. It was concluded, however, that better response did not translate into survival benefit (Goorin et al. 2003). The Cooperative Osteosarcoma Study Group study COSS-82 explored reduced intensity of preoperative chemotherapy and the salvage of poor responders (defined as less than 90 % tumor necrosis) and concluded that changing drugs for salvage failed to improve survival outcomes (Winkler et al. 1993). EURAMOS, a joint protocol of the world’s leading multi-institutional osteosarcoma groups (Children’s Oncology Group (COG), Cooperative Osteosarcoma Study Group (COSS), European Osteosarcoma Intergroup (EORTC/MRC), Scandinavian Sarcoma Group (SSG)), aimed to optimize the treatment of osteosarcoma patients through its collaboration. The EURAMOS-1 trial took into account the strong prognostic value of tumor response to preoperative chemotherapy and divided patients accordingly. Postoperative therapy was determined by the histological response of the tumor, with poor responders (defined as less than 90 % necrosis) stratified to receive intensified chemotherapy, while good responders were introduced to biological agents. Details of this study are described in Chap. 6. More recently, in cases of resectable osteosarcoma, the COG study AOST0331 sought to optimize treatment strategies based upon histological response to preoperative chemotherapy (www.clinicaltrials.gov, NCT00134030). In a recent meta-analysis of therapeutic regimens for localized high-grade osteosarcoma, it was concluded that the salvage of poor responders by changing drugs or intensifying treatment postoperatively does not prove to be efficacious (Anninga et al. 2011). It has also been shown that response to preoperative chemotherapy is more important than primary metastases in predicting survival for patients with osteosarcoma of the extremities (Bielack et al. 2002). Recent results of COG AOST0331/EURAMOS-1 have shown no benefit to intensified therapy for poor responders with localized osteosarcoma (http://www.ssg-org.net/wp-content/uploads/2011/03/EURAMOS-1-Poor-Response-Randomisation1.pdf).
8.2.3 Prognostic Indications of Histological Response in Ewing Sarcomas
As in osteosarcoma, poor histological response has been demonstrated to be an independent adverse prognostic factor for EFS in Ewing sarcoma patients (Bacci et al. 2004b). These patients exhibit drastic differences in relapse-free survival according to their histological response (Qureshi et al. 2013). Preoperative chemotherapeutic response has been reported to be the single-most predictive indicator of event-free survival in postoperative Ewing sarcoma patients (Wunder et al. 1998). In addition, histological response was determined to have the largest impact in the prediction of local recurrence of Ewing sarcoma after surgical treatment of the primary tumor, with central site of disease as a second independent predictive factor (Lin et al. 2007). Significant correlations between histological response to chemotherapy and primary tumor location, presence of metastases, and histological features relating to patient survival have been demonstrated (Coffin et al. 2005).
Also similar to what is seen in osteosarcoma, studies of the treatment of Ewing sarcoma patients based upon their histological response have yielded varying results. According to the French Society of Pediatric Oncology’s study, EW88, which used a threshold of 95 % tumor necrosis as a good response in localized Ewing sarcoma patients, further therapeutic trials were recommended based on histological response or tumor volume according to the method used for local control (Lopez Guerra et al. 2012; Oberlin et al. 2001). The German Cooperative Ewing Sarcoma Study, CESS 86, demonstrated poor histological response as a negative predictor of EFS. The study also noted that 52 % of patients survived after risk-adapted therapy (Paulussen et al. 2001a). It has since been demonstrated that the outcome of nonmetastatic Ewing sarcoma bone tumors is influenced by several variables in addition to histological response. It has been proposed that criteria to stratify Ewing sarcoma patients according to risk of relapse include all variables that show prognostic significance, such as gender, age, volume of tumor, type of local treatment, type of chemotherapy, the presence of distant recurrences, and serum LDH level, as opposed to being based on a single prognostic factor (Bacci et al. 2006a; Lopez Guerra et al. 2012).
8.2.4 Alternative Methods of Determining Histological Response
Recent advances in radiological technologies have led to the exploration of radiographic determinants for histological response to neoadjuvant chemotherapy. 18 F-fluorodeoxy-d-glucose (FDG)-positron emission tomography (PET) SUVmax is surging as a correlative predictor of histological response (Dutour et al. 2009; Kim et al. 2011). 18 F-FDG PET is a noninvasive imaging modality that predicts histological response to chemotherapy of various malignancies (Kim et al. 2011). 18 F-FDG PET SUVmax has been reported to correlate with histological response in osteosarcoma and may even predict response earlier than histological analysis (Dutour et al. 2009). It has also been reported that the relative responses in SUV from prechemotherapy (SUV1) to postchemotherapy (SUV2) as calculated by SUV(2:1) [(SUV1-SUV2)/SUV1] ≥0.5 and SUV2 ≤ 2.5 are related to favorable histological responses to chemotherapy, with sensitivity of SUV(2:1) at 0.5 and SUV2 at 2.5 of 93 % and 88 % and specificity at 88 % and 78 %, respectively (Kim et al. 2011). Metabolic tumor volume (MTV) measured by 18 F-FDG PET can also predict outcome of osteosarcoma of the extremities (Byun et al. 2013). MTV has been reported to be an independent predictor of metastasis in this group of patients, and the combination of MTV with histological response predicted survival more accurately than chemotherapeutic response alone (Byun et al. 2013). The change in SUVmax between baseline and post-treatment imaging is not significantly associated with histological response for either Ewing sarcoma or osteosarcoma. 18 F-FDG PET responses to neoadjuvant chemotherapy are different for Ewing sarcoma and osteosarcoma (Gaston et al. 2011). Observed differences between the groups include the overall MTV, likely secondary to a difference in the distribution of the injected 18F-FDG dose within the primary tumor. Therefore, a 50 % reduction in MTV, associated with favorable histological response in osteosarcoma, was found to incorrectly predict good responders in Ewing sarcoma. An increase of the Ewing sarcoma cutoff values to a 90 % MTV reduction was necessary for it to be associated with a favorable outcome (Gaston et al. 2011). These data suggest that response to chemotherapy as reflected by changes in PET characteristics should be interpreted differently in the two sarcoma types (Gaston et al. 2011).
8.3 Biological Markers
In addition to histological response induced by neoadjuvant chemotherapy, other prognostic indicators have emerged for malignant bone tumors. The presence of certain molecular or genetic features has been described to be an important prognostic marker, and new data arise regularly from preclinical studies with newly proposed markers such as those described below (Hingorani et al. 2013; Kobayashi et al. 2010; Limmahakhun et al. 2011; van Doorninck et al. 2010). The utilization of biomarkers and the ongoing discovery of new potential therapeutic targets have led researchers and clinicians one step closer in the approach toward patient risk stratification and personalized targeted therapeutic strategies. With the steady emergence of various molecules being touted as potential “biomarkers,” a set of guidelines, the REMARK criteria, has also been developed to elucidate their clinical usefulness. As a translational approach, biological markers have been categorized according to their utility and may be considered as either diagnostic, prognostic, or potential therapeutic targets (further discussed in Chap. 15).
8.3.1 The REMARK Criteria
“Personalized medicine,” involving the notion of individualized treatment plans according to the individual patient’s disease and outcome, is the ultimate goal for the treatment of patients with cancer including those with bone tumors. Biomarkers are being studied to guide decision-making toward a rational and directed approach. Along these lines, there has been an emergence of numerous proposed biomarkers. Without a standardized approach, however, which path to choose is unclear. Therefore, the precise definition of what qualifies as a “biomarker” has been developed, along with guidelines as to how that biomarker should be validated and applied in clinical protocols. A biomarker must provide a distinct risk:benefit ratio to enable clinical decision-making, be easily available, be cost effective, and be capable of being performed with available technologies.
Progressions in laboratory techniques and expansions of tumor tissue banking have led to more studies that include aims to explore potential biomarkers. However, conflicting results have often arisen. As a result, level of evidence (LOE) measures were created originally by the American Society of Clinical Oncology and have been further modified (Hayes et al. 1996; Simon et al. 2009). Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) guidelines were published by the National Cancer Institute in 2005 and recently updated in 2012. These guidelines require studies of biomarkers to meet specific criteria in order to be considered adequate, including clearly describing characteristics and treatment modalities for all patients and specimens involved, use of reproducible methodologies including disclosure of assay methods and detailed study design, and involving a clear biostatistical strategy. Data should be clearly reported, with analyses showing the relation of the marker to standard prognostic variables. The criteria also highlight the significance of reporting study limitations and implications for future research and clinical value (Altman et al. 2012; McShane et al. 2005).
8.3.2 Biological Markers in Osteosarcoma
Osteosarcoma is characterized by complex karyotypic abnormalities without chromosomal translocations or distinct patterns. Numerous genes have been shown to contribute to osteosarcoma tumorigenesis. Although continual progress has been made in detecting genetic alterations in osteosarcoma, to date no distinctive molecular marker has been proven to have better prognostic implications than the clinical markers which are currently employed. Nonetheless, the search continues to identify markers that may be utilized to guide therapeutic modifications. Some of the well-known biomarkers are discussed below.
DMP-1
Dentin Matrix Protein 1 (DMP-1) has been proposed to be a diagnostic marker of bone-forming tumors. It is highly expressed in osteocytes and found in dentin and bone. Expression has been reported in osteosarcoma, with negative stains in other bony neoplasms such as Ewing sarcoma, chondrosarcoma, leiomyosarcoma, fibrosarcoma, and giant cell tumor of bone (Kashima et al. 2013). As there are more definitive pathological techniques and stains to distinguish osteosarcomas, the role of DMP-1 remains unclear.
Oncogenes and Tumor Suppressor Genes
Several oncogenes and tumor suppressor genes have been implicated to have prognostic significance in osteosarcoma. The ongoing COG biology study, AOST06B1, examines the potential roles of RB/p53, ERBB2, MDM2, p16, and p21, LOH at 3q and 18q, the amplification or overexpression of C-sis, Gli, and C-fos, the presence of the SV-40T-antigen sequence, and Myc/Ras pathway (Martin et al. 2014; Patino-Garcia et al. 2003).
RB1
One of the earliest associations of tumor suppressor genes with osteosarcoma was found in patients who survived bilateral retinoblastoma. Loss of heterozygosity (LOH) of the retinoblastoma gene RB1 was identified to be associated with the development of osteosarcoma and can be found in a high percentage of patients with osteosarcoma. Studies investigating correlation of outcome with RB1 gene status have shown mixed results, and significant association has not been identified in more recent studies (Kong and Hansen 2009; Patino-Garcia et al. 2003).
p53 and MDM2
Osteosarcoma is one of the most common tumors associated with patients with Li–Fraumeni syndrome (Kong and Hansen 2009). As a result, cell cycle regulators such as p53 and its related proteins p16 and p21 are suspected to serve as prognostic indicators in osteosarcoma (Diller et al. 1990; Fu et al. 2013; McIntyre et al. 1994; Miller et al. 1996a; Scholz et al. 1992; Ueda et al. 1993). p53 has been highlighted as an effective biomarker for the prediction of survival in patients with osteosarcoma (Fu et al. 2013). Amplification of MDM2, a negative regulator of p53, has also been observed in metastatic or recurrent osteosarcomas and is thought to potentially play a role as an indirect pathway to p53 inactivation (Ladanyi et al. 1993a; Miller et al. 1996b; Patino-Garcia et al. 2003). Studies investigating the function of TP53 in projecting osteosarcoma outcomes have shown that the functional status of p53 is not associated with response to chemotherapy, but mutant p53 status has been reported to be associated with slight decrease in overall survival (Kong and Hansen 2009).
HER2/Neu
The proto-oncogene HER2/Neu, erbB-2, is a member of the epidermal growth factor receptor family and is variably expressed in osteosarcoma cells (Kong and Hansen 2009). Expression of HER2/erbB-2 has been reported to correlate with survival in osteosarcoma (Gorlick et al. 1999; Morris et al. 2001; Onda et al. 1996). Targeted therapy against erbB2, however, was proven to be ineffective for treatment of osteosarcoma (Kong and Hansen 2009).
LOH of 3q and 18q
Loss of heterozygosity in 3q and 18q has been frequently observed in osteosarcoma (Kong and Hansen 2009; Kruzelock et al. 1997). These locations may contain other tumor suppressor genes which have not yet been identified. They are currently both under review as potential prognostic indicators in the COG biology committee, although there is no published literature directly correlating loss of heterozygosity at either locus with patient outcomes.
Myc and Ras
A small percentage of osteosarcoma patients have been reported to have Myc amplification (Barrios et al. 1993; Ladanyi et al. 1993b). The amplification of C-myc is implicated in osteosarcoma pathogenesis, with studies demonstrating enhanced invasion characteristics of osteosarcoma cell lines with its amplification via the MEK-ERK pathway (Han et al. 2012). Amplification of c-myc has been correlated with a worse prognosis and decreased 3-year overall survival for osteosarcoma patients (Wu et al. 2012). Interestingly, it has also been reported that c-myc amplification occurs most frequently alongside of RB1 alterations, indicating that these may together play a role in the pathogenesis and progression of osteosarcoma (Ozaki et al. 1993). Ras gene mutations have been identified in several cancer types, but their role in osteosarcoma has not yet been determined (Antillon-Klussmann et al. 1995; Barrios et al. 1993; Nardeux et al. 1987).
C-sis and C-fos
C-sis encodes a chain of the platelet-derived growth factor, PDGF, and it has been identified as amplified in human osteosarcoma cells, as well as in spontaneous canine osteosarcomas (Graves et al. 1984; Kochevar et al. 1990). C-fos plays a role in normal bone metabolism, and osteosarcoma develops in transgenic mice which overexpress c-fos (Wu et al. 1990). Human osteosarcoma patients have been found to overexpress c-fos as well, though it is yet unclear whether or not it relates to prognosis (under study by COG).
Genes Related to Interactions with the Environment
Other potential prognostic indicators in osteosarcoma include the expression of genes related to environmental interactions, such as metalloproteinase (MMP), c-MET, insulin-like growth factor-1 (IGF-1), Ezrin, and Survivin. The metalloproteinases are thought to be responsible for lysing extracellular matrix proteins and therefore possibly play a role in invasion and metastasis (Kong and Hansen 2009). The c-met oncogene codes for hepatocyte growth factor (HGF) receptor. Stimulation of c-met induces cellular responses which include cell division, and overexpression of c-met has been correlated with poor outcomes in adult tumors. C-met is highly expressed in 60 % of human osteosarcoma, but it has not yet been studied as a prognostic indicator (Ferracini et al. 1995). Since peak incidence of osteosarcoma coincides with periods of rapid bone growth, it is possible that hormones which influence growth contribute to its development. Insulin-like growth factor 1 is regulated by growth hormone and has been demonstrated to function in the growth, turnover, and metabolism of normal bone. Osteosarcoma cell lines have displayed IGF-1 dependence in culture, and several drugs that inhibit growth hormone are being examined for the treatment of a variety of cancers (Burrow et al. 1998; Jentzsch et al. 2014; MacEwen et al. 2004). Ezrin (VIL2), an actin-cytoskeleton cross-linker, is thought to be essential for metastasis. In a limited sample of pediatric osteosarcoma patients, Ezrin expression was shown to be associated with favorable outcome (Kong and Hansen 2009; Zhang et al. 2014). Survivin (BIRC5) is a member of the Inhibitor of Apoptosis (IAP) family of proteins. It binds caspase-3 and -7 and was found to be expressed in osteosarcoma tumors but not in normal tissues, and its expression has been correlated with increased malignancy and metastasis (Hingorani et al. 2013; Kong and Hansen 2009).
Chromosomes
Markers associated with chromosome function such as telomerase and tumor cell ploidy are also being studied as potential prognostic indicators for osteosarcoma. Telomerase serves as a regulator of the number of replications a cell may undergo (Wen et al. 2002), and it is hypothesized that it may play a role in the progression of tumors. Telomerase expression has been negatively correlated with prognosis in several adult cancers including osteosaracoma (Kong and Hansen 2009; Nakashima et al. 2003; Sanders et al. 2004). Though ploidy has been previously investigated as a prognostic indicator in prior osteosarcoma studies, it is being re-examined in the context of current clinical trials (Bauer et al. 1989; Gebhardt et al. 1990).
Drug Resistance Genes
The MDR1 gene encodes P-glycoprotein, a transmembrane protein responsible for the efflux of numerous chemotherapeutic agents including doxorubicin. Multidrug resistance protein (MRP) is a family member of P-glycoprotein and is responsible for efflux of doxorubicin and etoposide (Grant et al. 1994; Loe et al. 1996). Methotrexate transport and metabolism via the dihydrofolate reductase enzyme has been shown to play a significant role in treatment of osteosarcoma. Increased expression of glutathione S-transferase P1 (GSTP1), a phase II detoxification enzyme, has been associated with significantly higher relapse rate and a worse clinical outcome in osteosarcoma (Kong and Hansen 2009). It was found to be upregulated in osteosarcoma cells when treated with doxorubicin or cisplatin.
Bone Turnover Markers
Markers for bone turnover have also been studied as potential prognostic indicators for osteosarcoma. Serological markers like alkaline phosphatase have been reported to be predictive of time to recurrence for osteosarcoma patients (Bacci et al. 2002b). Higher levels are associated with earlier relapses of the disease (Kong and Hansen 2009). It has been proposed that alkaline phosphatase may be useful in the monitoring and assessment of efficacy of therapy in pediatric osteosarcoma (Ambroszkiewicz et al. 2006, 2010a, b).
Indications of bone marrow recovery after chemotherapy may be of prognostic value in osteosarcoma as well. The time for recovery of lymphocytes has been reported to be a significant indicator of outcomes in osteosarcoma patients. Using a threshold absolute lymphocyte count of ≥ 800 cells/μl on day 14, researchers were able to distinguish significant differences in overall survival based upon “early” versus “late” lymphocyte recovery, with a reported 5 year OS of 92.3 % versus 33.3 %, respectively. Risk stratification and subsequent therapy based on the threshold absolute lymphocyte count on day 14 may thus be a rational strategy that could be tested in a clinical trial (Moore et al. 2010).
8.3.3 Biological Markers in Ewing Sarcoma
Several studies of prognostic biomarkers in Ewing sarcoma have yielded the emergence of four main categories which adhere to REMARK criteria, including EWSR1 translocation type, cell cycle proteins, copy number alterations (CNAs), and subclinical disease measurement (Pinto et al. 2011; van Maldegem et al. 2012; Wagner et al. 2012), among others.
EWS Translocation
A translocation involving the EWSR1 gene on chromosome 22 is the molecular hallmark of Ewing sarcomas. Typically, the 5′ of EWSR1 is fused to the 3′ of an ETS gene family member, most commonly FLI1 on chromosome 11, t(11;22)(q24;q12). The EWS-FL1 fusion protein was the first sarcoma gene to be cloned and represents a dominant oncogene. This translocation has been further categorized into the more frequently occurring Type 1, involving exon 6, and Type 2, involving exon 5. The remaining minority of cases are comprised of the 3′ translocation companion with various ETS family genes (Sankar and Lessnick 2011), including ERG on chromosome 21 in 10 %, EWS-ETV1 (<1 %), EWS-ETV4 (<1 %), EWS-FEV (<1 %).
The fusion type may have prognostic relevance in Ewing sarcoma (Halperin 2011). Correlations between prognosis and fusion type were initially reported from retrospective studies conducted in the late 1990s, with reports of patients with type-1 fusion having significantly improved overall survival compared to those with other fusion types. In other reports, lower relapse rates were associated with type-1 fusion in patients with localized disease. Both Euro-EWING and COG subsequently evaluated outcomes and fusion status, with results that were not supportive of the original findings (Le Deley et al. 2010; Shukla et al. 2013). In the Euro-EWING trial, there was no difference exhibited in the distribution of sex, age, tumor volume, tumor site, disease extension, or histological response between the four fusion type groups. The study concluded that EWS fusion type was not prognostically significant for risk of disease progression or relapse (Le Deley et al. 2010).
Cell Cycle Proteins
Cell cycle regulation markers have been shown to have clinical significance in Ewing sarcomas (Lopez-Guerrero et al. 2011). Genetic alterations in the RB pathway have been described in Ewing sarcoma, including deletions of both RB1 and CDKN2A (INK4A/ARF) (Huang et al. 2005; Kovar et al. 1997; Mackintosh et al. 2012). Numerous retrospective studies have shown correlations between patient outcome and alterations in CDKN2A (Kovar et al. 1997). In 2013, the COG Ewing Sarcoma Biology Committee found strong evidence to support CDKN2A loss as a strong negative prognostic marker (Shukla et al. 2013).
The mutational status of p53 has also been assessed in retrospective studies as a potential prognostic biomarker in Ewing sarcoma (Yang et al. 2014). p53 overexpression has been associated with advanced disease at time of diagnosis, poorer treatment response, and inferior overall survival, independent of site, local treatment, or extent of tumor necrosis (Abudu et al. 1999; Yang et al. 2014). High p53 expression has also been reported to be the strongest prognostic factor correlating with decreased overall survival (de Alava et al. 2000). p53 mutations and/or CDKN2A deletions are associated with poor response to chemotherapy (Huang et al. 2005). A significant association exists between increased p53 expression and metastatic disease and poorer progression-free survival and disease-specific survival in patients with localized disease (Lopez-Guerrero et al. 2011).
Copy Number Alterations
Copy number alterations (CNAs) and genomic instability have been repeatedly reported in Ewing sarcoma (Jahromi et al. 2012; Shukla et al. 2013). The most commonly reported alterations in Ewing sarcoma are trisomy 8, trisomy 12, and gain of 1q (Shukla et al. 2013). Alterations that correlated with outcome are 1p36.3 loss (Hattinger et al. 1999), 1q21-q22 gain (Armengol et al. 1997; Kullendorff et al. 1999; Mackintosh et al. 2012; Tarkkanen et al. 1999), 6p21.1 gain (Tarkkanen et al. 1999), 8 gain (Armengol et al. 1997; Jahromi et al. 2012; Tarkkanen et al. 1999; Zielenska et al. 2001), 12 gain (Armengol et al. 1997; Tarkkanen et al. 1999; Zielenska et al. 2001), 16q loss (Jahromi et al. 2012; Ozaki et al. 2001), 20 gain (Jahromi et al. 2012; Roberts et al. 2008), or other combinations (Ferreira et al. 2008; Jahromi et al. 2012; Kullendorff et al. 1999; Ozaki et al. 2001; Roberts et al. 2008; Savola et al. 2009; Zielenska et al. 2001). Independent studies have identified chromosomal copy number alterations as accepted prognostic indicators. The COG Ewing Sarcoma Biology Committee recommends that tumor and germline DNA be collected from all Ewing sarcoma patients registered on future therapeutic studies so that CNAs and other genetic mutations can be evaluated as prognostic and predictive biomarkers (Shukla et al. 2013). The COG has also discussed prospective incorporation of copy number alterations in its upcoming relapsed/refractory Ewing sarcoma trial (Shukla et al. 2013).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
