Molecular feature-based classification of retroperitoneal liposarcoma: a prospective cohort study

Retroperitoneal liposarcoma (or RPLS for short) is a rare type of cancer that forms a tumor deep in the abdomen. Although it can be treated through surgery, results vary widely, and the cancer often returns. Determining which RPLS tumours are likely to respond well to treatment, and which are more aggressive, can be challenging. This is partly because many of the tumors appear similar under a microscope, making it difficult to identify those that might require more intense treatment.

For some types of cancer, researchers have started to overcome this problem by identifying molecular markers – specific genes, proteins and other molecules – that are unique to tumor cells. This allows for a better understanding of tumor behavior and helps pinpoint which treatments are most likely to be successful. However, the relevant molecular markers for RPLS remain unknown, preventing this precision medicine approach from being applied to this type of cancer.

To help close this gap, Xiao et al. studied the genetic activity of tumors from 88 RPLS patients to look for activity associated with more aggressive tumors. Analysis revealed two distinct groups of tumors. One group showed high activity of a gene known as LEP, which is involved in metabolism. Tumors with high LEP activity were less aggressive, with patients surviving longer and requiring fewer surgeries. The other tumor group showed high activity of a gene known as PTTG1, which plays a role in cell division and DNA repair. When overactive, these processes can make tumors grow and spread more quickly. In this group, tumors were more likely to return after surgery. Based on these findings, Xiao et al. developed a simple test using the two genes to classify tumors. When tested in a larger group of 241 patients, this method successfully sorted tumors into the two groups with more than 99% accuracy.

The findings of Xiao et al. suggest that the activity of LEP and PTTG1 could potentially be used as a molecular marker for identifying RPLS patients that might benefit from more intensive treatments. However, further research is required to confirm how reliable these markers are before they can be used to diagnose patients in the clinic.

Retroperitoneal liposarcoma (RPLS) is a soft tissue sarcoma originating in the retroperitoneum with an insidious onset. Traditional surgical resection has been regarded as a primary and curable treatment strategy of RPLS for the past 50 years (Ecker et al., 2016). However, the anatomical complexity and biological properties of sarcoma brought great difficulty in achieving microscopically margin-negative resection, leading to a high postoperative recurrence rate in RPLS patients. During the past decade, scientists tried to improve the postoperative survival of RPLS patients by personalized surgical resection and neoadjuvant/adjuvant therapies, but the effect was not satisfactory (Littau et al., 2021; Gronchi et al., 2015; Gronchi et al., 2009; Gronchi and Pollock, 2013; Pisters, 2009).

Recently, precision medicine greatly enriched the therapeutic approaches and reformed the clinical decision-making chain of tumor diagnosis and treatment, prolonging the median survival of main tumor types 2–10 times (Kam et al., 2021; Zeng and Jin, 2022; Alifrangis et al., 2019; Frese et al., 2021). Biomarker-based patient stratification and targeted therapy together make up the kernel of precision medicine, which is intrinsically based on the molecular profiling of cancers. However, our knowledge of the molecular features of RPLS is limited, and few clinically applicable molecular biomarkers and targeted drugs are available for RPLS treatment. Only sporadic molecules such as CDK4 (Pilotti et al., 2000), MDM2 (Binh et al., 2005), AURK4 (Yen et al., 2019), and CCNDBP1 (Yang et al., 2021) have been reported as prognostic and diagnostic biomarkers, but these biomarkers were poorly represented and verified. Therefore, it is crucial to reveal the molecular landscape of RPLS and explore a feasible classification for its diagnosis and treatment.

Here, we conducted a comprehensive investigation into the molecular characteristics of RPLS through the delineation of the largest gene expression landscape ever assembled for this rare disease entity. By identifying both RPLS-specific genes and prognostic biomarkers, we unveiled their intricate relationships with clinical parameters. Our findings revealed the existence of two distinct molecular subtypes within all RPLS patients, characterized by diverse pathological compositions, enriched signaling pathways, and varying clinical outcomes. This highlights the limitations of relying solely on traditional pathological classification for surgical decision-making in certain cases where patients exhibit favorable histological features but poor prognoses, emphasizing the pivotal role of molecular subtyping in guiding individualized treatment strategies and enhancing patient management. To facilitate practical application in clinical settings, we developed a simplified RPLS classification system based on key biomarkers (LEP and PTTG1) representative of each subtype. Notably, this classification scheme was validated in a larger cohort of RPLS patients through immunohistochemistry (IHC) assays (Figure 1), laying the groundwork for precise surgical interventions guided by molecular insights in the realm of RPLS treatment.

Figure 1

Download asset Open asset

Baseline characteristics were shown in Table 1. Of 329 RPLS patients, 88 in training cohort and 241 in validation cohort. No statistically significant differences were found in the age, sex, pathology, surgery times, tumor size, and multilocation between the two cohorts (p>0.05). In validation cohort, the IHC scores of LEP and PTTG1 were 1.62 (0.820) and 0.830 (0.75), respectively.

Cell cycle, DNA damage and repair, and metabolism are dysregulated in RPLS

To reveal the general molecular features of RPLS compared to noncancerous adipose tissues, we first recruited eight RPLS patients and collected paired tumor and normal tissues for DEG analysis. A total of 1354 DEGs, 554 upregulated and 800 downregulated, were identified (Figure 2A and B). To assess the underlying pathways of RPLS, GSEAs were performed for those DEGs. We found that proliferation-associated pathways, such as mitotic spindle, E2F target, G2/M checkpoint, and separation of sister chromatids, were mainly enriched in tumors, while metabolism-related pathways, such as bile acid metabolism, heme and fatty acid metabolism, and integration of energy metabolism, were enriched in normal controls (Figure 2C).

Figure 2

Download asset Open asset

Then, we collected another 80 samples to investigate the molecular heterogeneity of RPLS. Gene expression profiles showed 918 and 3244 genes associated with OS and DFS, respectively. Among 497 candidate genes associated with both OS and DFS, 83 of them also overlapped with DEGs (Figure 2B). Functional annotation (GO, KEGG, and enrichWP) demonstrated that cell cycle, DNA damage and repair, and metabolism-related pathways were significantly enriched (Figure 2D), suggesting these signaling pathways were dysregulated in RPLS.

RPLS subgroups based on molecular features show different clinical outcomes

To evaluate heterogeneous molecular clustering characteristics in RPLS, ssGSEA emerged as a widely adopted method for computing the enrichment level of specific biological signaling pathways for each sample based on gene expression data. This aids in gaining insights into the overall activity level of signaling pathways. Here, we scored each sample on the dysregulated pathways by ssGSEA and divided RPLS patients into two subgroups (Figure 3A). Subgroup 1 (G1) showed better OS and DFS compared to subgroup 2 (G2) (Figure 3B and C). G1 displayed elevated ssGSEA scores associated with metabolism, whereas G2 exhibited heightened ssGSEA scores linked to cell cycle and DNA damage and repair (Figure 3D). These features suggested that effective monitoring of the prognosis of RPLS patients can be achieved based on the activation status of specific pathways. We also evaluated the clinical features and immune infiltration levels of those samples. The results showed that G1 had a lower aggressive pathological composition ratio, MDM2 and Ki67 expression, larger tumor size, and higher tumor microenvironment (TME) level compared to G2 (Figure 3E, G, H; Figure 3—figure supplement 1A and B). Surgery times for G1 were also tended to decrease (Figure 3F). Taken together, the above results indicated that RPLS subgroups based on molecular features showed distinct clinical features and clinical outcomes.

Figure 3 with 1 supplement see all

Download asset Open asset

A simplified RPLS classification strategy derived from RPLS G1/G2 subgroups

To explore representative biomarkers for different RPLS subgroups, we performed a DEG analysis between G1 and G2. There were 1258 genes downregulated, among which 112 of them indicated good prognosis (protective genes). Correspondingly, 754 genes were upregulated, and 28 of them indicated poor prognosis (aggressive genes) (Figure 3—figure supplement 1C and D). Enrichment analysis suggested that those DEGs were also associated with cell cycle regulation and metabolism (Figure 3—figure supplement 1E), which was consistent with previous results (Figure 2C).

To develop a simplified RPLS clustering based on DEGs, we adopted NMF and t-SNE for a reclassification of those patients. The results showed RPLS patients were also divided into two clusters (Figure 4A and Figure 3—figure supplement 1F and G). We then annotated the samples of two clusters by ssGSEA and found Cluster1 (C1) was related to metabolic processes, and Cluster2 (C2) was mainly related to the processes of cell cycle and DNA damage and repair (Figure 4—figure supplement 1A). Also, C1 showed better OS and DFS, lower pathological composition ratio and MDM2 expression, and fewer surgery times (Figure 4B–F). Lower Ki67 expression and larger tumor size were observed in C2 (Figure 4—figure supplement 1B and C). Interestingly, the biological annotations of the C1/C2 classification were greatly consistent with G1/G2. Therefore, a simplified RPLS classification strategy derived from RPLS subgroups was provisionally established.

Figure 4 with 1 supplement see all

Download asset Open asset

Development of a dichotomous RPLS classification model

For NMF classification of RPLS patients, LEP and PTTG1 were identified as representative biomarkers of C1 and C2, respectively (Figure 5A). The selection criteria for identifying LEP and PTTG1 as biomarkers involved selecting prognostic genes that were highly expressed in C1 and C2, respectively, and achieved the highest AUC value in distinguishing the two RPLS groups. We aimed to replicate the RPLS classification of C1 and C2 by integrating these two biomarkers with the assistance of machine learning algorithms, and this two-gene panel achieved promising results (logistic, AUC = 0.995; SVM, AUC = 0.997; RF, AUC = 1.000; Figure 5B). Also, a linear negative correlation between LEP and PTTG1 expression was detected (Figure 5C). Considering the enhanced interpretability and generalization of linear models, we adopted the results of LR for subsequent analysis (risk values = 2.182 × PTTG1−2.204×LEP). The patients marked as high-risk (Cluster_H) exhibited worse OS and DFS than those marked as low-risk (Cluster_L) (Figure 5D and E). Dysregulated pathways, such as DNA repair and cell cycle regulation, were enriched in Cluster_H (Figure 5F; Figure 4—figure supplement 1D), and Cluster_H presented more aggressive pathological composition ratio, higher MDM2 levels, and marginally increased in surgery times than Cluster_L (Figure 5G–I). Similarly, the Cluster_H showed higher Ki67 levels and smaller tumor sizes (Figure 4—figure supplement 1E and F). Moreover, a Sankey diagram was drawn to show the correlation among G1/G2, C1/C2, and Cluster_L/H. Cluster_L/H was well matched to C1/C2 and G1/G2, suggesting LEP and PTTG1 were promising biomarkers for a dichotomous RPLS classification (Figure 5J). To ensure the broader applicability of LEP and PTTG1 as classification biomarkers, we performed an independent validation using an external liposarcoma cohort (GSE30929). This dataset was selected due to its relevance to RPLS (N=63, 45%) and the availability of distant recurrence-free survival (DRFS) outcomes, aligning with the clinical focus of our study. Applying our established LR, we found that the high-risk (V) group exhibited significantly worse DRFS compared to the low-risk (V) group (Figure 5—figure supplement 1A and B), and the high-risk group (V) demonstrated a higher proportion of high-grade histology (Figure 5—figure supplement 1C and D), consistent with our previous findings. These results validate the robustness and generalizability of our risk stratification model across distinct liposarcoma cohorts. The external dataset’s alignment with our findings underscores the potential of LEP and PTTG1 as reproducible biomarkers for prognosis and therapeutic stratification in liposarcoma.

Figure 5 with 1 supplement see all

Download asset Open asset

Validation of the dichotomous RPLS classification in another 241 RPLS patients

To validate LEP and PTTG1 as biomarkers for a dichotomous RPLS classification, we performed IHC staining of two biomarkers in validation cohort. The representative images of LEP and PTTG1 with different expression levels were shown in Figure 6A and B. The IHC scores were integrated with the previously fitted coefficients to evaluate the prognosis of RPLS patients (risk values = 2.182×PTTG1_IHC-2.204×LEP_IHC). The cutoff value of validation cohort is the median of the risk value. The high-risk group had worse OS and DFS (Figure 6C and D), along with more surgery times and more aggressive pathological composition ratio (Figure 7A and B), but the difference in tumor size was not observed between the two groups (Figure 7C). Then, we constructed visual nomograms for a precise survival prediction of RPLS patients by combining the risk score with clinical features. The predictive abilities of the 1-, 2-, and 3-year OS (Figure 7D–F) and DFS (Figure 7—figure supplement 1A–C) were 0.743–0.788. Together, we proposed a simple and clinically applicable molecular classification strategy for RPLS patients.

Figure 6

Download asset Open asset

Figure 7 with 1 supplement see all

Download asset Open asset

Here, we divided RPLS patients into two subgroups based on cell cycle, DNA damage and repair, and metabolism-related pathways. G1 was annotated as metabolism-active, which exhibited high ssGSEA scores on metabolism-associated pathways, while G2 showed high ssGSEA scores on cell cycle and DNA damage and repair with a high Ki67 and MDM2 level.

G2 had more aggressive molecular features and worse clinical outcomes compared to G1, in accordance with previously reported tumor classification (Lindskrog et al., 2021; Yu et al., 2021; Zhang et al., 2022). In fact, the Cancer Genome Atlas Research Network. Electronic address: elizabeth.demicco@sinaihealthsystem.ca and Cancer Genome Atlas Research Network, 2017, has integrated SCNA and DNA methylation to divide dedifferentiated liposarcoma into two subtypes (S1 and S2), the unfavorable cluster was characterized as JUN amplified (an oncogene that promotes proliferation and metastasis) and lower inferred fraction of immature dendritic cells. However, the patients in the Cancer Genome Atlas Research Network. Electronic address: elizabeth.demicco@sinaihealthsystem.ca and Cancer Genome Atlas Research Network, 2017, were of complex origin (mixed limbs, trunk, and retroperitoneum), providing limited guidance for RPLS molecular classification. Here, we reported the first clinically applicable RPLS molecular classification based on RNA sequencing and IHC validation cohorts.

To facilitate the clinical application, we constructed a simplified RPLS molecular classification derived from the original cell cycle/metabolism subgroups. By the NMF algorithm, we identified LEP and PTTG1 as representative biomarkers for each subtype. A model based on IHC staining of LEP and PTTG1 successfully approximated the original dichotomous RPLS classification in biological features and survival outcomes. LEP is an important regulator of basal metabolism and food intake, which is considered a linkage between metabolism and the immune system (Jiménez-Cortegana et al., 2021). Although LEP-based targeting therapies have not yet been fully applied, LEP has already been identified as a potent metabolic reprogramming agent to support antitumor responses in aggressive melanomas (Waldman et al., 2020; de la Cruz-Merino et al., 2019; Rivadeneira et al., 2019). In addition, LEP improves the immunotherapeutic effects by regulating innate and adaptive immune responses via increasing the cytotoxicity of NK cells (Francisco et al., 2018), stimulating the proliferation of T/B cells (Francisco et al., 2018; Bernotiene et al., 2006), and activating DC cells (Hu et al., 2019). Those reported roles of LEP provided a good mechanism explanation on the features of metabolism pathway-enriched, better prognosis, and higher TME level of metabolism subgroup (LEP+). In contrast, PTTG1 acts as a regulator of sister chromatid separation during cell division under physiological conditions (Zou et al., 1999), which is closely linked to genetic instability, aneuploidy, tumor progression, invasion, and metastasis (Heaney et al., 2000; Ramaswamy et al., 2003; Kim et al., 2005; Yu et al., 2003; Teveroni et al., 2021; Romero et al., 2001). PTTG1 also regulates the cell cycle and the transactivation of growth factors as an initiator and promoter of tumorigenesis (Zou et al., 1999; Mora-Santos et al., 2013; McCabe et al., 2002; Ishikawa et al., 2001; Hamid et al., 2005). Overexpressing PTTG1 was correlated with worse prognosis in tumors, such as ovarian cancer (Parte et al., 2019), cervical cancer (Guo et al., 2019b), renal cell carcinoma (Tian et al., 2022), and colorectal cancer (Heaney et al., 2000). Therefore, the biological functions of PTTG1 provided a good mechanism explanation of the pathway-enriched cell cycle/DNA damage and repair-associated, worse prognosis, and more aggressive pathological composition ratio in the cell cycle subgroup (PTTG1+).

Data supporting the conclusions of this article are presented within the article and its supplementary files. Source data of Figure 2–5, Figure 3—figure supplement 1 and Figure 4—figure supplement 1 are available in Supplementary files 1, 2 and 4. Source data of Figure 6–7 and Figure 7—figure supplement 1 are available in Supplementary files 3. Source data of Figure 5—figure supplement 1 is available in GSE30929. The sequencing data (Supplementary files 1–4) of retroperitoneal liposarcoma have been deposited at the Open Archive for Miscellaneous Data (OMIX) database of China National Center for Bioinformation (CNCB) under the accession number OMIX002786. The data of the external validation cohort (GSE30929) was downloaded from GEO (https://www.ncbi.nlm.nih.gov/geo/). This paper does not report the original code.

1. 1. Xiao MM

   (2023) CNCB

   ID OMIX002786. Whole genome sequencing of retroperitoneal liposarcoma.

   https://ngdc.cncb.ac.cn/omix/release/OMIX002786
2. 1. Gobble RM
   2. Qin LX
   3. Brill ER
   4. Angeles CV

   (2011) NCBI Gene Expression Omnibus

   ID GSE30929. Whole-transcript expression data for liposarcoma.

   https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE30929
3. 1. Xiao MM

   (2023) CNCB

   ID PRJCA014351. Whole genome sequencing of retroperitoneal liposarcoma.

   https://ngdc.cncb.ac.cn/omix/release/PRJCA014351

1. 1. Alifrangis C
   2. McGovern U
   3. Freeman A
   4. Powles T
   5. Linch M

   (2019) Molecular and histopathology directed therapy for advanced bladder cancer

   Nature Reviews. Urology 16:465–483.

   https://doi.org/10.1038/s41585-019-0208-0

   • PubMed
   • Google Scholar
2. 1. Bernotiene E
   2. Palmer G
   3. Gabay C

   (2006) The role of leptin in innate and adaptive immune responses

   Arthritis Research & Therapy 8:e2004.

   https://doi.org/10.1186/ar2004

   • PubMed
   • Google Scholar
3. 1. Binh MBN
   2. Sastre-Garau X
   3. Guillou L
   4. de Pinieux G
   5. Terrier P
   6. Lagacé R
   7. Aurias A
   8. Hostein I
   9. Coindre JM

   (2005) MDM2 and CDK4 immunostainings are useful adjuncts in diagnosing well-differentiated and dedifferentiated liposarcoma subtypes: a comparative analysis of 559 soft tissue neoplasms with genetic data

   The American Journal of Surgical Pathology 29:1340–1347.

   https://doi.org/10.1097/01.pas.0000170343.09562.39

   • PubMed
   • Google Scholar
4. 1. Cancer Genome Atlas Research Network. Electronic address: elizabeth.demicco@sinaihealthsystem.ca
   2. Cancer Genome Atlas Research Network

   (2017) Comprehensive and integrated genomic characterization of adult soft tissue sarcomas

   Cell 171:950–965.

   https://doi.org/10.1016/j.cell.2017.10.014

   • PubMed
   • Google Scholar
5. 1. de la Cruz-Merino L
   2. Palazón-Carrión N
   3. Henao-Carrasco F
   4. Nogales-Fernández E
   5. Álamo-de la Gala M
   6. Vallejo-Benítez A
   7. Chiesa M
   8. Sánchez-Margalet V
   9. GEICAM (Spanish Breast Cancer Research Group) and GÉTICA (Spanish Group for Cancer Immuno-Biotherapy)

   (2019) New horizons in breast cancer: the promise of immunotherapy

   Clinical & Translational Oncology 21:117–125.

   https://doi.org/10.1007/s12094-018-1907-3

   • PubMed
   • Google Scholar
6. 1. Ecker BL
   2. Peters MG
   3. McMillan MT
   4. Sinnamon AJ
   5. Zhang PJ
   6. Fraker DL
   7. Levin WP
   8. Roses RE
   9. Karakousis GC

   (2016) Preoperative radiotherapy in the management of retroperitoneal liposarcoma

   The British Journal of Surgery 103:1839–1846.

   https://doi.org/10.1002/bjs.10305

   • PubMed
   • Google Scholar
7. 1. Francisco V
   2. Pino J
   3. Campos-Cabaleiro V
   4. Ruiz-Fernández C
   5. Mera A
   6. Gonzalez-Gay MA
   7. Gómez R
   8. Gualillo O

   (2018) Obesity, fat mass and immune system: role for leptin

   Frontiers in Physiology 9:640.

   https://doi.org/10.3389/fphys.2018.00640

   • PubMed
   • Google Scholar
8. 1. Frese KK
   2. Simpson KL
   3. Dive C

   (2021) Small cell lung cancer enters the era of precision medicine

   Cancer Cell 39:297–299.

   https://doi.org/10.1016/j.ccell.2021.02.002

   • PubMed
   • Google Scholar
9. 1. Gronchi A
   2. Lo Vullo S
   3. Fiore M
   4. Mussi C
   5. Stacchiotti S
   6. Collini P
   7. Lozza L
   8. Pennacchioli E
   9. Mariani L
   10. Casali PG

   (2009) Aggressive surgical policies in a retrospectively reviewed single-institution case series of retroperitoneal soft tissue sarcoma patients

   Journal of Clinical Oncology 27:24–30.

   https://doi.org/10.1200/JCO.2008.17.8871

   • PubMed
   • Google Scholar
10. 1. Gronchi A
    2. Pollock RE

    (2013) Quality of local treatment or biology of the tumor: which are the trump cards for loco-regional control of retroperitoneal sarcoma?

    Annals of Surgical Oncology 20:2111–2113.

    https://doi.org/10.1245/s10434-013-2971-0

    • PubMed
    • Google Scholar
11. 1. Gronchi A
    2. Miceli R
    3. Allard MA
    4. Callegaro D
    5. Le Péchoux C
    6. Fiore M
    7. Honoré C
    8. Sanfilippo R
    9. Coppola S
    10. Stacchiotti S
    11. Terrier P
    12. Casali PG
    13. Le Cesne A
    14. Mariani L
    15. Colombo C
    16. Bonvalot S

    (2015) Personalizing the approach to retroperitoneal soft tissue sarcoma: histology-specific patterns of failure and postrelapse outcome after primary extended resection

    Annals of Surgical Oncology 22:1447–1454.

    https://doi.org/10.1245/s10434-014-4130-7

    • PubMed
    • Google Scholar
12. 1. Guo L
    2. Chen G
    3. Zhang W
    4. Zhou L
    5. Xiao T
    6. Di X
    7. Wang Y
    8. Feng L
    9. Zhang K

    (2019a) A high-risk luminal A dominant breast cancer subtype with increased mobility

    Breast Cancer Research and Treatment 175:459–472.

    https://doi.org/10.1007/s10549-019-05135-w

    • PubMed
    • Google Scholar
13. 1. Guo X-C
    2. Li L
    3. Gao Z-H
    4. Zhou H-W
    5. Li J
    6. Wang Q-Q

    (2019b) The long non-coding RNA PTTG3P promotes growth and metastasis of cervical cancer through PTTG1

    Aging 11:1333–1341.

    https://doi.org/10.18632/aging.101830

    • PubMed
    • Google Scholar
14. 1. Hamid T
    2. Malik MT
    3. Kakar SS

    (2005) Ectopic expression of PTTG1/securin promotes tumorigenesis in human embryonic kidney cells

    Molecular Cancer 4:3.

    https://doi.org/10.1186/1476-4598-4-3

    • PubMed
    • Google Scholar
15. 1. Heaney AP
    2. Singson R
    3. McCabe CJ
    4. Nelson V
    5. Nakashima M
    6. Melmed S

    (2000) Expression of pituitary-tumour transforming gene in colorectal tumours

    Lancet 355:716–719.

    https://doi.org/10.1016/S0140-6736(99)10238-1

    • PubMed
    • Google Scholar
16. 1. Hu W
    2. Wang G
    3. Huang D
    4. Sui M
    5. Xu Y

    (2019) Cancer immunotherapy based on natural killer cells: current progress and new opportunities

    Frontiers in Immunology 10:1205.

    https://doi.org/10.3389/fimmu.2019.01205

    • PubMed
    • Google Scholar
17. 1. Ishikawa H
    2. Heaney AP
    3. Yu R
    4. Horwitz GA
    5. Melmed S

    (2001) Human pituitary tumor-transforming gene induces angiogenesis

    The Journal of Clinical Endocrinology and Metabolism 86:867–874.

    https://doi.org/10.1210/jcem.86.2.7184

    • PubMed
    • Google Scholar
18. 1. Jiménez-Cortegana C
    2. López-Saavedra A
    3. Sánchez-Jiménez F
    4. Pérez-Pérez A
    5. Castiñeiras J
    6. Virizuela-Echaburu JA
    7. de la Cruz-Merino L
    8. Sánchez-Margalet V

    (2021) Leptin, both bad and good actor in cancer

    Biomolecules 11:913.

    https://doi.org/10.3390/biom11060913

    • PubMed
    • Google Scholar
19. 1. Kam AE
    2. Masood A
    3. Shroff RT

    (2021) Current and emerging therapies for advanced biliary tract cancers

    The Lancet. Gastroenterology & Hepatology 6:956–969.

    https://doi.org/10.1016/S2468-1253(21)00171-0

    • PubMed
    • Google Scholar
20. 1. Kim D
    2. Pemberton H
    3. Stratford AL
    4. Buelaert K
    5. Watkinson JC
    6. Lopes V
    7. Franklyn JA
    8. McCabe CJ

    (2005) Pituitary tumour transforming gene (PTTG) induces genetic instability in thyroid cells

    Oncogene 24:4861–4866.

    https://doi.org/10.1038/sj.onc.1208659

    • PubMed
    • Google Scholar
21. 1. Li X
    2. Bu F
    3. Ma S
    4. Cananzi F
    5. Zhao Y
    6. Xiao M
    7. Min L
    8. Luo C

    (2022) The Janus-faced role of TRPM2-S in retroperitoneal liposarcoma via increasing ROS levels

    Cell Communication and Signaling 20:128.

    https://doi.org/10.1186/s12964-022-00873-9

    • PubMed
    • Google Scholar
22. 1. Lindskrog SV
    2. Prip F
    3. Lamy P
    4. Taber A
    5. Groeneveld CS
    6. Birkenkamp-Demtröder K
    7. Jensen JB
    8. Strandgaard T
    9. Nordentoft I
    10. Christensen E
    11. Sokac M
    12. Birkbak NJ
    13. Maretty L
    14. Hermann GG
    15. Petersen AC
    16. Weyerer V
    17. Grimm MO
    18. Horstmann M
    19. Sjödahl G
    20. Höglund M
    21. Steiniche T
    22. Mogensen K
    23. de Reyniès A
    24. Nawroth R
    25. Jordan B
    26. Lin X
    27. Dragicevic D
    28. Ward DG
    29. Goel A
    30. Hurst CD
    31. Raman JD
    32. Warrick JI
    33. Segersten U
    34. Sikic D
    35. van Kessel KEM
    36. Maurer T
    37. Meeks JJ
    38. DeGraff DJ
    39. Bryan RT
    40. Knowles MA
    41. Simic T
    42. Hartmann A
    43. Zwarthoff EC
    44. Malmström PU
    45. Malats N
    46. Real FX
    47. Dyrskjøt L

    (2021) An integrated multi-omics analysis identifies prognostic molecular subtypes of non-muscle-invasive bladder cancer

    Nature Communications 12:2301.

    https://doi.org/10.1038/s41467-021-22465-w

    • PubMed
    • Google Scholar
23. 1. Littau MJ
    2. Kulshrestha S
    3. Bunn C
    4. Agnew S
    5. Sweigert P
    6. Luchette FA
    7. Baker MS

    (2021) The importance of the margin of resection and radiotherapy in retroperitoneal liposarcoma

    American Journal of Surgery 221:554–560.

    https://doi.org/10.1016/j.amjsurg.2020.11.041

    • PubMed
    • Google Scholar
24. 1. McCabe CJ
    2. Boelaert K
    3. Tannahill LA
    4. Heaney AP
    5. Stratford AL
    6. Khaira JS
    7. Hussain S
    8. Sheppard MC
    9. Franklyn JA
    10. Gittoes NJL

    (2002) Vascular endothelial growth factor, its receptor KDR/Flk-1, and pituitary tumor transforming gene in pituitary tumors

    The Journal of Clinical Endocrinology and Metabolism 87:4238–4244.

    https://doi.org/10.1210/jc.2002-020309

    • PubMed
    • Google Scholar
25. 1. McShane LM
    2. Altman DG
    3. Sauerbrei W
    4. Taube SE
    5. Gion M
    6. Clark GM
    7. Statistics Subcommittee of the NCI-EORTC Working Group on Cancer Diagnostics

    (2005) REporting recommendations for tumour MARKer prognostic studies (REMARK)

    British Journal of Cancer 93:387–391.

    https://doi.org/10.1038/sj.bjc.6602678

    • PubMed
    • Google Scholar
26. 1. Mora-Santos M
    2. Castilla C
    3. Herrero-Ruiz J
    4. Giráldez S
    5. Limón-Mortés MC
    6. Sáez C
    7. Japón MÁ
    8. Tortolero M
    9. Romero F

    (2013) A single mutation in Securin induces chromosomal instability and enhances cell invasion

    European Journal of Cancer 49:500–510.

    https://doi.org/10.1016/j.ejca.2012.06.024

    • PubMed
    • Google Scholar
27. 1. Parte S
    2. Virant-Klun I
    3. Patankar M
    4. Batra SK
    5. Straughn A
    6. Kakar SS

    (2019) PTTG1: a unique regulator of stem/cancer stem cells in the ovary and ovarian cancer

    Stem Cell Reviews and Reports 15:866–879.

    https://doi.org/10.1007/s12015-019-09911-5

    • PubMed
    • Google Scholar
28. 1. Pilotti S
    2. Della Torre G
    3. Mezzelani A
    4. Tamborini E
    5. Azzarelli A
    6. Sozzi G
    7. Pierotti MA

    (2000) The expression of MDM2/CDK4 gene product in the differential diagnosis of well differentiated liposarcoma and large deep-seated lipoma

    British Journal of Cancer 82:1271–1275.

    https://doi.org/10.1054/bjoc.1999.1090

    • PubMed
    • Google Scholar
29. 1. Pisters PWT

    (2009) Resection of some -- but not all -- clinically uninvolved adjacent viscera as part of surgery for retroperitoneal soft tissue sarcomas

    Journal of Clinical Oncology 27:6–8.

    https://doi.org/10.1200/JCO.2008.18.7138

    • PubMed
    • Google Scholar
30. 1. Ramaswamy S
    2. Ross KN
    3. Lander ES
    4. Golub TR

    (2003) A molecular signature of metastasis in primary solid tumors

    Nature Genetics 33:49–54.

    https://doi.org/10.1038/ng1060

    • PubMed
    • Google Scholar
31. 1. Rivadeneira DB
    2. DePeaux K
    3. Wang Y
    4. Kulkarni A
    5. Tabib T
    6. Menk AV
    7. Sampath P
    8. Lafyatis R
    9. Ferris RL
    10. Sarkar SN
    11. Thorne SH
    12. Delgoffe GM

    (2019) Oncolytic viruses engineered to enforce leptin expression reprogram tumor-infiltrating T cell metabolism and promote tumor clearance

    Immunity 51:548–560.

    https://doi.org/10.1016/j.immuni.2019.07.003

    • PubMed
    • Google Scholar
32. 1. Romero F
    2. Multon MC
    3. Ramos-Morales F
    4. Domínguez A
    5. Bernal JA
    6. Pintor-Toro JA
    7. Tortolero M

    (2001) Human securin, hPTTG, is associated with Ku heterodimer, the regulatory subunit of the DNA-dependent protein kinase

    Nucleic Acids Research 29:1300–1307.

    https://doi.org/10.1093/nar/29.6.1300

    • PubMed
    • Google Scholar
33. 1. Teveroni E
    2. Di Nicuolo F
    3. Bianchetti G
    4. Epstein AL
    5. Grande G
    6. Maulucci G
    7. De Spirito M
    8. Pontecorvi A
    9. Milardi D
    10. Mancini F

    (2021) Nuclear localization of PTTG1 promotes migration and invasion of seminoma tumor through activation of MMP-2

    Cancers 13:212.

    https://doi.org/10.3390/cancers13020212

    • PubMed
    • Google Scholar
34. 1. Tian X
    2. Xu WH
    3. Xu FJ
    4. Li H
    5. Anwaier A
    6. Wang HK
    7. Wan FN
    8. Zhu Y
    9. Cao DL
    10. Zhu YP
    11. Shi GH
    12. Qu YY
    13. Zhang HL
    14. Ye DW

    (2022) Identification of prognostic biomarkers in papillary renal cell carcinoma and PTTG1 may serve as a biomarker for predicting immunotherapy response

    Annals of Medicine 54:211–226.

    https://doi.org/10.1080/07853890.2021.2011956

    • PubMed
    • Google Scholar
35. 1. Waldman AD
    2. Fritz JM
    3. Lenardo MJ

    (2020) A guide to cancer immunotherapy: from T cell basic science to clinical practice

    Nature Reviews. Immunology 20:651–668.

    https://doi.org/10.1038/s41577-020-0306-5

    • PubMed
    • Google Scholar
36. 1. Yang L
    2. Wu Z
    3. Sun W
    4. Luo P
    5. Chen S
    6. Chen Y
    7. Yan W
    8. Li Y
    9. Wang C

    (2021) CCNDBP1, a prognostic marker regulated by DNA methylation, inhibits aggressive behavior in dedifferentiated liposarcoma via repressing epithelial mesenchymal transition

    Frontiers in Oncology 11:687012.

    https://doi.org/10.3389/fonc.2021.687012

    • PubMed
    • Google Scholar
37. 1. Yen CC
    2. Chen SC
    3. Hung GY
    4. Wu PK
    5. Chua WY
    6. Lin YC
    7. Yen CH
    8. Chen YC
    9. Wang JY
    10. Yang MH
    11. Chao Y
    12. Chang MC
    13. Chen WM

    (2019) Expression profile‑driven discovery of AURKA as a treatment target for liposarcoma

    International Journal of Oncology 55:938–948.

    https://doi.org/10.3892/ijo.2019.4861

    • PubMed
    • Google Scholar
38. 1. Yu R
    2. Lu W
    3. Chen J
    4. McCabe CJ
    5. Melmed S

    (2003) Overexpressed pituitary tumor-transforming gene causes aneuploidy in live human cells

    Endocrinology 144:4991–4998.

    https://doi.org/10.1210/en.2003-0305

    • PubMed
    • Google Scholar
39. 1. Yu Z
    2. Deng P
    3. Chen Y
    4. Liu S
    5. Chen J
    6. Yang Z
    7. Chen J
    8. Fan X
    9. Wang P
    10. Cai Z
    11. Wang Y
    12. Hu P
    13. Lin D
    14. Xiao R
    15. Zou Y
    16. Huang Y
    17. Yu Q
    18. Lan P
    19. Tan J
    20. Wu X

    (2021) Inhibition of the PLK1-coupled cell cycle machinery overcomes resistance to oxaliplatin in colorectal cancer

    Advanced Science 8:e2100759.

    https://doi.org/10.1002/advs.202100759

    • PubMed
    • Google Scholar
40. 1. Zeng Y
    2. Jin RU

    (2022) Molecular pathogenesis, targeted therapies, and future perspectives for gastric cancer

    Seminars in Cancer Biology 86:566–582.

    https://doi.org/10.1016/j.semcancer.2021.12.004

    • PubMed
    • Google Scholar
41. 1. Zhang F
    2. Zhang Q
    3. Zhu J
    4. Yao B
    5. Ma C
    6. Qiao N
    7. He S
    8. Ye Z
    9. Wang Y
    10. Han R
    11. Feng J
    12. Wang Y
    13. Qin Z
    14. Ma Z
    15. Li K
    16. Zhang Y
    17. Tian S
    18. Chen Z
    19. Tan S
    20. Wu Y
    21. Ran P
    22. Wang Y
    23. Ding C
    24. Zhao Y

    (2022) Integrated proteogenomic characterization across major histological types of pituitary neuroendocrine tumors

    Cell Research 32:1047–1067.

    https://doi.org/10.1038/s41422-022-00736-5

    • PubMed
    • Google Scholar
42. 1. Zou H
    2. McGarry TJ
    3. Bernal T
    4. Kirschner MW

    (1999) Identification of a vertebrate sister-chromatid separation inhibitor involved in transformation and tumorigenesis

    Science 285:418–422.

    https://doi.org/10.1126/science.285.5426.418

    • PubMed
    • Google Scholar

Author details

1. Mengmeng Xiao

   1. Department of Retroperitoneal Tumor Surgery, Peking University People’s Hospital, Beijing, China
   2. Department of Retroperitoneal Tumor Surgery, Peking University International Hospital, Beijing, China

   Contribution

   Data curation, Formal analysis, Methodology, Writing - original draft, Writing – review and editing

   Contributed equally with

   Xiangji Li and Fanqin Bu

   Competing interests

   No competing interests declared

2. Xiangji Li

   1. Department of Retroperitoneal Tumor Surgery, Peking University International Hospital, Beijing, China
   2. Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, State Key Laboratory for Digestive Health, National Clinical Research Center for Digestive Disease, Beijing Digestive Disease Center, Beijing Key Laboratory for Precancerous Lesion of Digestive Disease, Beijing, China

   Contribution

   Data curation, Formal analysis, Investigation, Methodology, Writing - original draft

   Contributed equally with

   Mengmeng Xiao and Fanqin Bu

   Competing interests

   No competing interests declared

3. Fanqin Bu

   Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, State Key Laboratory for Digestive Health, National Clinical Research Center for Digestive Disease, Beijing Digestive Disease Center, Beijing Key Laboratory for Precancerous Lesion of Digestive Disease, Beijing, China

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing – review and editing

   Contributed equally with

   Mengmeng Xiao and Xiangji Li

   Competing interests

   No competing interests declared

4. Shixiang Ma

   Department of Retroperitoneal Tumor Surgery, Peking University International Hospital, Beijing, China

   Contribution

   Formal analysis, Methodology

   Competing interests

   No competing interests declared

5. Xiaohan Yang

   Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, State Key Laboratory for Digestive Health, National Clinical Research Center for Digestive Disease, Beijing Digestive Disease Center, Beijing Key Laboratory for Precancerous Lesion of Digestive Disease, Beijing, China

   Contribution

   Investigation, Methodology

   Competing interests

   No competing interests declared

6. Jun Chen

   Department of Retroperitoneal Tumor Surgery, Peking University International Hospital, Beijing, China

   Contribution

   Formal analysis

   Competing interests

   No competing interests declared

7. Yu Zhao

   Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, State Key Laboratory for Digestive Health, National Clinical Research Center for Digestive Disease, Beijing Digestive Disease Center, Beijing Key Laboratory for Precancerous Lesion of Digestive Disease, Beijing, China

   Contribution

   Investigation

   Competing interests

   No competing interests declared

8. Ferdinando Cananzi

   Department of Biomedical Sciences, Humanitas University, Milan, Italy

   Contribution

   Investigation

   Competing interests

   No competing interests declared

9. Chenghua Luo

   1. Department of Retroperitoneal Tumor Surgery, Peking University People’s Hospital, Beijing, China
   2. Department of Retroperitoneal Tumor Surgery, Peking University International Hospital, Beijing, China

   Contribution

   Conceptualization, Supervision, Methodology, Writing – review and editing

   For correspondence

   pkuihlch@163.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5939-7888

10. Li Min

    Department of Gastroenterology, Beijing Friendship Hospital, Capital Medical University, State Key Laboratory for Digestive Health, National Clinical Research Center for Digestive Disease, Beijing Digestive Disease Center, Beijing Key Laboratory for Precancerous Lesion of Digestive Disease, Beijing, China

    Contribution

    Conceptualization, Resources, Supervision, Writing – review and editing

    For correspondence

    minli@ccmu.edu.cn

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0001-9595-5536

• 171

  views

• 13

  downloads

• 0

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.