Running title:TP73 deregulation in cancer
The expression levels of TP73, a member of the TP53 family, has been reported de-regulated in a wide variety of cancers and linked to patients´ outcome. The fact that TP73 encodes a complex number of isoforms (TAp73 and ΔTAp73) with opposing functions and the cross-talk with other members of the family (TP53 and TP63) makes it difficult to determine its clinical relevance. Here, we discuss the current data available in the literature that provide evidence on the role of TP73 variants as prognostic markers in cancer. To date, most of the studies that evaluate the status levels of TP73 isoforms have been based on limited-size series. Despite this limitation, these publications highlight the correlation between high levels of the oncogenic forms and failure to respond to chemotherapy and/or shorter survival. We also argue on tumor derived-microvesicles/biological material present in the blood stream as the biological material to easily check the expression status of TP73 variants with prognosis purposes. Finally, we emphasize the need for studies to evaluate the significance of combining the deregulation of various members of the TP53 family in order to define patient outcome or their responsiveness to specific therapies.
The TP73 gene has been accompanied by controversy since its identification as a TP53 homologue (Jost et al., 1997).The high structural homology with TP53 and its capability to induce cell cycle arrest and apoptosis (Scian et al., 2008) initially pointed to a new tumour suppressor gene. Conversely, no mutations were observed inTP73, but its over-expression in many tumour types (Melino et al., 2002), there was no cancer phenotype in the knock-out mice (Yang et al., 2000), and some authors even described oncogenic properties for TP73 (Stiewe et al., 2002a; Zaika et al., 2002). The description from 2000 onwards of different amino-terminal isoforms and the release of the specific knock-out mice variant was the initial point to unravel the role of the different isoforms in cancer (Fillippovich et al., 2001; Tomasini et al., 2008; Wilhelm et al., 2010).
TP73 translates into different variants with opposing functions: TAp73 and ΔEx2p73, ΔEx2/3p73, ΔNp73 and ΔN´p73. These isoforms can in addition be processed at the COOH-terminal (Fig. 1). TAp73 variants contain the NH2-terminal transactivation domain and mimic TP53 activity in experimental systems (Jost et al., 1997; Domínguez et al., 2006a). ΔEx2p73, ΔEx2/3p73, ΔNp73, and ΔN´p73 (known as ∆TAp73) lack part or the whole amino terminal transactivation domain. The Ex2p73 and Ex2/3p73 isoforms lack exon 2 and exons 2 and 3, respectively (Kaghad et al., 1997; Sherr and Weber, 2000; Domínguez et al., 2006a). Np73 is generated from an alternative promoter at intron 3 and contains a newly described 3´ exon (Domínguez et al., 2006; Wilhelm et al., 2010). The N´p73 variant generates from the first promoter, but includes 198 bp from the 3´ exon. Translation of the N´p73 transcript starts at the 3´ exon, and result in a protein that is indistinguishable from Np73 (Ishimoto et al., 2002a; Domínguez et al., 2006). Thus, Ex2p73, Ex2/3p73, Np73, and N´p73 are transactivation-deficient variants that fail to induce apoptosis and cell cycle arrest by acting as direct competitors for the DNA binding-sites and/or by hetero-duplex formation with TP53 and the transactivation-competent TAp73 variants, inhibiting their tumour-suppressor properties (Fillippovich et al., 2001; Stiewe et al., 2002a; Zaika et al., 2002; Domínguez et al., 2006). Furthermore, ∆Np73 isoforms show a longer half-life than the TAp73 variants, and therefore, even if all NH2-terminal isoforms are identically expressed, ∆Np73 isoforms are predominant due to their higher stability. Remarkably, a self-regulatory feedback loop exists, since TAp73 forms and TP53 transactivate ∆Np73 forms their by direct binding to the second promoter (Grob et al., 2001). Additionally, ∆Np73 variants induce the expression of their own set of genes (Lin et al., 2009) through a second transactivation domain. These ∆Np73 variant effectors could also be responsible for their oncogenic potential (Vilgelm et al., 2008; Steder M et al., 2013; Dulloo et al., 2015; Fernandez-Alonso R et al., 2015; Agostini M et al., 2016; Landré V et al., 2016).
The isoform-specific knock-out animal models has also clarified the specific role of each variant in carcinogenesis. Mice defective for TAp73 variants show an increased incidence of spontaneous and carcinogen-induced tumours, probably due to the genomic instability associated with enhanced aneuploidy (Tomasini et al., 2008). In contrast, no evidence of tumour development exists in ∆Np73 knock-out mice (Wilhelm et al., 2010).
Thus, the literature describes a tumour suppressive role for TAp73 variants and an oncogenic role for ∆TAp73 forms. Nevertheless, there are few data in the literature about the impact of the aberrant levels of specific TP73 amino isoforms on cancer patient outcome and its plausible clinical application. We discuss here the relevance of these publications and their translational potential.
TP73 is essential for brain development (Gonzalez-Cano L et al., 2016; Niklison-Chirou MV et al., 2013). While ΔNp73 plays a key anti-apoptotic role during postmitotic neuron maturation, TAp73 is expressed in adult neurons and related to apoptosis and cell cycle arrest (Pozniak et al., 2000; Killick et al., 2011). The coordinated function of TP73, TP63 and TP53 seems a crucial factor in neuro-development (Fatt MP et al., 2014). Additionally, several reports have shown an association between deregulated TP73 expressionand brain tumour formation.
Medulloblastoma, the most common malignant brain tumour of childhood, shows a different molecular profile from other brain tumours. It is frequently diagnosed between the age of 5 and 9, and in adults, comprises fewer than 2% of central nervous system malignancies (Remke et al., 2011).
Few groups have studied the contribution of TP73 to medulloblastoma formation and progression. The study by Castellino et al. (2007) of 34 patients detected overexpression of TAp73 and ΔNp73 in medulloblastoma, although tumours showed higher ΔNp73 than TAp73 levels. The patients with high levels of TAp73 showed better disease-free survival and overall survival (P=0.052 and P=0.07, respectively). The analysis of 29 medulloblastoma samples by Zitterbart et al. (2007) showed that patients with ΔNp73 over-expression survived for a shorter time than those with low ΔNp73 (P=0.026). Therefore, the contribution of the specific TP73 isoforms to the malignant phenotype in this neoplasia clearly reveals the tumour-suppressor role of TAp73 variants and the oncogenic role of ∆Np73 ones.
Neuroblastoma, a solid and malignant tumour, is the second most common extracranial tumour in patients younger than 15 (7% of malignancies). This tumour presents with remarkable biological and clinical heterogeneity and its progression depends on stage, age at diagnosis, and numerous molecular parameters (Rufini et al., 2011). The fact that the TP73 chromosomal region is frequently deleted in these patients shows its involvement in neuroblastoma development.In 1999 Romani and co-workers reported that the TP73 was expressed in approximately 50% of 130 neuroblastomas with no significant correlation with age, clinical stage, or 3-year overall survival (Romani et al., 1999), indicating that TP73 is not essential in the pathogenesis of neuroblastoma. By 2002, the expression and the clinical significance of the specific variant ΔNp73 was investigated in the same set of patients. ΔNp73 was detected in 30% of tumours, more frequently in children aged under 12 months and with advanced disease . The patients expressing both isoforms showed a variable ratio of TA/ΔN, a likely consequence of the heterogeneity of neuroblastoma. Additionally, ΔNp73 over-expression correlated independently with reduced progression-free survival (P<0.001) and with reduced overall survival (P<0.001), supporting its possible value as a strong prognostic factor in this disease (Casciano et al., 2002).
Glioma is the most common and aggressive primary brain tumour in adults, accounting for more than 50% of all primary intracranial cancers. Low-grade gliomas are well-differentiated (not anaplastic) and have better prognosis than high–grade gliomas. Even though the relation between TP73 expression and neural tumorigenesis is evident, only one prospective study to evaluate its involvement in gliomagenesis and its prognostic relevance has been released. In 2006, Wager and co-workers reported that all TP73 isoforms were over-expressed in 51 tumour samples evaluated, but only ΔEx2/3p73 and ΔEx2p73 seem to be prognostic factors for overall survival (P=0.07 and P=0.017, respectively). TAp73 and ΔNp73 did not appear to be involved in the survival rates and, consequently, in the development of this neoplasia (Wager et al., 2006).
All these studies support that the deregulation of TP73 plays a critical role in the pathogenesis of brain tumours and highlight the significance of specific TP73 isoforms as plausible and useful prognostic markers in these types of cancers.
Gynaecological cancers account for 8% of malignancies in women and 45% of all genital cancers (Jemal et al., 2011). Cancer of the cervix is the most common gynaecological cancer, followed by cancers of the uterus and the ovary.
A first study in a series of 100 ovarian cancer (Concin et al., 2004) showed that expression of ΔTAp73 isoforms distinguishes ovarian cancer patients from healthy controls. ΔN´p73 showed the strongest and most prevalent tumour-specific increased expression, with a marked upregulation in most cases. For the remaining variants, a small percentage of subjects showed upregulation of ΔNp73 and TAp73, downregulation of ∆Ex2/3p73, and no differences for ∆Ex2p73. Furthermore, co-upregulation of TAp73 and ΔTAp73 isoforms was found in tumours when compared with normal tissues. No statistical correlation between TP73 variant expression levels and patient survival was found, although a trend was observed for poor prognosis associated with high expression of ΔN`p73 and ΔNp73 isoforms (P=0.11). In a follow-up study within the same series extended to 122 ovarian cancer patients (Concin et al., 2005), the authors reported that high expression levels of ΔNp73 and ΔN´p73 correlated unfavourably with disease-free and overall survival (P=0.048 and P=0.005, respectively). Moreover, patients with poor prognosis were linked to chemotherapeutic failure and mutant TP53(P=0.048). Further statistical analysis noted that ΔTAp73 isoforms were independent prognostic markers (P=0.048).
A study published in 2006 reported that the expression of several TP73 forms associated with a favourable or unfavourable response to radiotherapy and, with prognosis of cervical squamous cell carcinoma (SCC) (Liu et al., 2006). Over-expression of both TAp73 and ΔNp73 was found in the vast majority of the 117 SCC cases studied. Subjects with higher levels of ΔNp73 tended to have lower levels of the TAp73 variants. This expression ratio was related to radio-resistance and recurrence of the disease (P=0.001 and P=0.012, respectively) and, with poor outcome of patients. A recent study by Zhu and co-workers supports these findings in SCC, since high expression levels of ΔNp73 may indicate an unfavourable prognosis in individuals at early stage of the disease (Zhu et al, 2015).
In breast cancer, a study analysed the expression levels of TP73 isoforms in 60 breast tumours and normal counterpart tissue and observed that the overexpression of both ΔNp73 and ΔEx2/3 associated with parameters associated with a poor prognosis, such as vascular invasion and oestrogen-negative receptors (Domínguez et al., 2006). Lately, a phase II clinical trial of platinum monotherapy in patients diagnosed with triple-negative breast cancer did not reveal a role of TP73 as a marker of response (Isakoff et al., 2015). This study focused specifically on the ratio ∆Np63/TAp73, thus, it cannot be ruled out the prognosis value of ∆TAp73 variants in this context.
The most relevant fact in gynaecological malignancies is the association of over-expression of ΔTAp73 isoforms with an unfavourable outcome due to the involvement of these variants in drug-resistance processes and tumour aggressiveness.
Hepatocellular carcinoma (HCC) is the most common type of liver cancer, whose prognosis depends on tumour size and staging. HCC usually develops in the setting of chronic liver disease, particularly in patients with chronic hepatitis B and C.
Muller et al. (2005) reported the putative role of ∆Np73 as a novel prognostic marker in HCC. Authors observed the overexpression of this variant in this cancer type and its correlation with a worse prognosis in 84 HCC patients. Subjects with high levels of ΔNp73 (37% of the cases) showed significantly shorter survival times than those whose tumours were ΔNp73-negative (P < 0.005). Later on, the same group (Schuster A et al, 2010) also reported that a ΔNp73 target gene signature is of prognostic relevance in HCC.
Lung cancer is characterized by its poor outcome, with an overall survival rate of about 12% (Ott and Geiser, 2012). Approximately 95 percent of all lung cancers are classified as either small cell lung cancer (SCLC) or non-small cell lung cancer (NSCLC). This distinction is essential for staging, treatment, and prognosis. Other cell types comprise about 5 percent of malignancies arising in the lung.
Uramoto et al (2004) pointed to the involvement of TP73 isoforms in the outcome of these patients. The study reinforced the role of ΔNp73 as an oncogenic variant and its role as a likely prognostic marker in this disease. Upregulation of ΔNp73 was found in 58 (35%) of the 132 patients enrolled in this study and directly associated with tumour stage (P=0.04). Furthermore, the authors also reported that positive expression of ΔNp73 was, along with pathological status, an independent factor for predicting poor prognosis (P<0.0001) (Uramoto et al., 2004). However, TAp73 expression strongly predicted good patient outcome (Amelio et al., 2015)
Colon cancer is one of the most common human malignancies and the second cause of cancer-related death worldwide.
It has been described the involvement of TP73 in the development of this tumour type and in the prediction of patient outcome. Sun et al (2002) reported that patients with TP73-positive tumours showed shorter survival than those with TP73-negative ones (P<0.05). The authors evaluated in 221 colorectal cancer patients total levels of TP73, but they did not take into consideration the different variants. It was later that the first study accounting for the status of all TP73 isoforms in colon cancer was released. This report (Domínguez et al. 2006) found the co-over-expression of TAp73 and ΔTAp73 (ΔEx2p73, ΔEx2/3p73 and ΔEx2/3p73) isoforms in 113 patients with colorectal cancer. Additionally, a direct statistical correlation was observed among the expression levels of all TP73 forms. The analysis of the pathological data revealed an association between upregulation of ΔTAp73 isoformsand presence of polyps, vascular invasion, and lymph node metastasis. The most relevant correlation observed was between over-expression of ∆Ex2/3p73 and of ∆Np73 and advanced tumour stage. A follow-up with the same series of patients (77 patients out of the initial 113) provided us with very relevant findings (Soldevilla et al., 2011). Interestingly, the over-expression of the two oncogenic isoforms that we found previously associated with advanced tumour stage, ∆Ex2/3 and ∆Np73, was significantly associated with shortened overall survival (P=0.038and P=0.06, respectively). We also observed positive correlation between upregulation of ΔTAp73 variants and of genes associated with drug resistance processes, such as HMGB1 and ABCB1 (P<0.05 and P=0.06, respectively). Our data indicated that genes associated with drug resistance could be the effectors of the oncogenic functions of ΔTAp73 forms. Notably, the over-expression of HMGB1 and ABCB1 was also strongly associated with poor prognosis (P=0.04 and P=0.03, respectively) and had an independent relationship with overall survival, in addition to tumour stage, in a multivariate analysis (p=0.008) (Soldevilla et al., 2011). Our data clearly showed the oncogenic potential of ΔTAp73 isoforms both in vitro and in vivo and the potential use of the aberrant expression levels of these variants and/or their effector genes in clinical practice as prognostic markers. Interestingly, our group also observed that the amount of ΔNp73 packaged in tumor-derived exosomes correlated with colon cancer patients’ survival (Soldevilla et al., 2014). Thus, the measurement of its levels through a non-invasive method could be closer to its clinical translation with a prognostic goal.
A publication by Toumi et al. (2010) supports these findings. Although the authors did not analyse the levels of specific TP73 isoforms, they detected their total expression. Toumi and co-workers observed that TP73 expression was markedly increased from normal mucosa (26%) to metastasis (97%) using a cohort of 204 patients with colorectal cancer. In addition, expression of TP73 correlated with shorter survival periods (P< 0.05).
Gastric and Oesophageal Carcinoma
The first study reporting data on the status of the ∆Np73 variant in gastric and oesophageal tumours was published in 2010. In this serie of 185 patients, Vilgelm et al. (2010) observed through an immunohistochemistry approach that this isoform was more significantly expressed in tumour epithelial cells than in normal mucosa. Importantly, those patients with a gastric tumour harbouring high levels of nuclear ∆Np73 showed shorter survival (P=0.005) than those with negative or weak expression. Similar results were obtained from patients with oesophageal or gastro-oesophageal junction cancer (P=0.009). The multivariate analysis revealed that ∆Np73 was not an independent prognosis factor.
In contrast, a recent report using 200 cases of oesophageal squamous cell carcinoma (Chen et al., 2015) supports a role of TP73 as a prognosis marker. In this work, a general antibody against p73 which recognizes a wide range of isoforms was used, thus these results are not comparable with those from Vilgelm and collaborators.
Head and Neck Squamous Cell Cancer
Although Faridoni-Laurens et al. (2001) initially described downregulation of general TP73 in head and neck squamous cell cancer (HNSCC), later their analysis using immunoblotting and RT-PCR reported that both TAp73 and ∆TAp73 isoforms were upregulated in HNSCC samples more than in normal tissue (Faridoni-Laurens et al., 2008). Specifically, ∆Np73 was the predominant and significantly over-expressed variant in the tumour (P=0.003). Although the upregulation of TAp73, ∆Ex2p73 and ∆Ex2-3p73 did not correlate with survival, a trend was observed between patients with low levels of ∆Np73 and better overall survival (P=0.107).
∆Np73 was detected by Rizzo et al. (2004) in 96.4% of 71 cases of M0, M1, M2, M4, M5 and M6 acute myeloid leukaemia (AML) and in 32% of cases of PML-RARA-positive M3 AML. The authors concluded that this specific variant could be involved in this neoplasia and differentiated a biological subset of AML, namely acute promyelocytic leukaemia. In this sense, it has been just released a work which reports that a higher ∆Np73/TAp73 ratio in patients with acute promyelocytic leukaemia associates with free-disease and overall survival and with treatment response (Lucena-Araujo et al., 2015).
Meier and co-workers found, in the T-lineage phenotype of childhood acute lymphoblastic leukaemia (ALL), total TP73 mRNA higher in patients than in controls, with ∆TAp73 forms contributing much more (88%) than TAp73 ones (12%). In addition, the authors emphasised that these isoforms may play a role in cell resistance to DNA-damaging drugs in children at initial diagnosis of T-ALL, since the ∆Ex2p73 form was most abundantly expressed in daunorubicin-resistant cells (Meier et al., 2006).
In B-cell chronic lymphocytic leukaemia (B-CLL), over-expression of TAp73 and ∆Np73 isoforms was observed in the vast majority of the 51 leukaemia cells, when compared with normal cells (Leupin et al., 2004).
Finally, a recent study has described the lack of TAp73 and ∆TAp73 isoforms in lymphoma patients but their upregulation in those with lymphoid leukemia (Hamdy et al., 2016). This data support the fact that the expression levels of TP73 variants may reflect the differences in biology between these malignancies and may be used as a specific marker.
Only one work regarding the involvement of TP73 in Wilms´ tumors has been published (Song et al., 2016). In this report the authors observed a significantly higher expression of TP73 in patients than controls, which is associated with shorter survival rates. More remarkable, TP73 was detected in the preoperative venous blood of these patients and showed a high diagnostic potential. Thus, in this tumor type the aberrant expression of TP73 may be used as a prognostic marker but also it can be used with diagnosis purposes through a non-invasive method for the patient.
The Role of TP73 in Other Tumours
The expression levels of TP73 isoforms in other tumour types have also been evaluated. Although deregulation of specific variants has been reported in certain cancers, their roles in prognosis are unknown due to a lack of follow-up of the patient series. Thus, Tuve and collaborators described in 2004 the over-expression of ∆EX2p73 and ∆Ex2/3p73 in melanoma metastasis, with ∆Np73 predominant in benign nevi (Tuve et al., 2004). They studied a series of 8 benign melanocytic nevi, 8 primary melanomas, and 19 melanoma metastases. Likewise, ∆Ex2/3p73 was found over-expressed in 7 uveal melanoma patients (Kilic et al., 2008). Although a role could be attributed to ∆Ex2p73 and ∆Ex2/3p73 in the development and progression of melanoma, the series size analysed in these studies is too small to get any clear conclusions. More recently, overexpression of TP73 has been reported in a series of 35 cutaneous melanoma samples but unfortunately, the authors did not further analyse the different specific variants (Ganzetti et al., 2015).
In vulval squamous cancer, it seems that ∆Ex2p73 also plays an important role: its upregulation was observed in high-grade premalignant lesions and tumours, with its levels increasing with the degree of neoplasm (O’Nions et al., 2001).
In 2004, Park et al. (2004) described a low expression of TAp73 in half of 12 osteosarcomas evaluated. As they mentioned no other TP73 isoforms, this fact and the few samples evaluated make the real involvement of TAp73 in the development of this neoplasia vague.
In the late 1990s, the analysis of TP73 expression levels in prostate cancer produced contradictory results. Takahashi and co-workers described an association between high levels of TP73 and tumour growth in 106 prostate tumours (Takahashi et al., 1998). However, Yokomizo´s group did not observe any difference in expression between 27 primary prostate cancers and their counterpart normal tissue (Yokomizo et al., 1999). It was not until 2005 that the status of the different isoforms was evaluated in this type of cancer. Thus, Guan and Chen(2005) reported that TAp73 and ∆Np73 were upregulated in 72.7% and 60%, respectively, of 33 prostate tumours. Interestingly, only the over-expression of ∆Np73 correlated with the tumour Gleason score (P=0.037), suggesting the potential role of this variant in prostate cancer progression.
Controversial results were also found regarding thyroid neoplasms. Ferru and collaborators observed downregulation of TAp73 and ∆Np73 in 60 adenomas and in differentiated carcinomas (Ferru et al., 2006), while upregulation of ∆Np73 and its correlation with carcinoma progression was reported in 223 thyroid tumours (Ito et al., 2006).
In bladder cancer, the expression of TP73 was reported to be undetectable or low in 68% of the 154 tumour tissues analysed and correlated inversely with invasiveness and cancer progression (P<0.0001) (Puig et al., 2003). In this publication, the only one for this tumour type, the role of the specific TP73 variant was not examined. Thus, although the case number was sufficient for the study to attain significance, it is difficult to infer the real role of this gene in this cancer.
Something similar occurs for pancreatic adenocarcinoma. In this case the over-expression of TP73 in 45.6% of the samples inversely linked with some clinico-pathological parameters, specifically in cystic adenocarcinomas (Ito et al., 2001). Recently, Thakur and co-workers have described the deletion of the TP73 locus in 86% of a series of 14 patients diagnosed with pancreatic ductal adenocarcinoma (Thakur et al., 2016). In a mouse model the authors observed that the specific loss of TAp73 associated with more aggressive tumours and shorter survival rates. This effect was mediated by the activation of the TFG-β signalling pathway. Interestingly, De la Fuente et al. (2015) have reported that inhibition of ITCH, an ubiquitin ligase which controls TP73 levels, associated to a better response to gemcitabine in mice with pancreatic cancer. Thus, although no conclusive results have been reported yet in this tumour type, these preliminary results point to the TP73 signalling as an interesting pathway to be further explored.
Finally, Chen et al. (2003) showed that upregulation of TP73 was also involved in the development of human buccal squamous-cell carcinoma and Weber´s and Choi´s groups found that it was involved in head and neck squamous tumorigenesis (Choi et al., 2002; Weber et al., 2002).
The deregulated expression of TP73 was also found in hilar cholangiocarcinoma (Zhao W et al., 2014) and in malignant parotid gland tumors (Yong et al., 2014), but the low number of cases and the fact that no single TP73 variants were analysed make it difficult to assign a role of TP73 in these malignancies.
In summary, there is evidence in different tumour types to support the view that TP73 isoforms have a function at various stages of the carcinogenesis process. Some of these studies are based on small patient series, so the results need to be confirmed in larger series. The follow-up of these series will provide us with valuable information on the real involvement of TP73 isoforms in certain tumours.
The members of the TP53 Family as Prognostic Markers in Cancer
Although a complex functional relationship has been reported among the different members of the TP53 family, few studies have demonstrated the importance of combining the altered expression or mutation of different members in the prognosis of cancer patients.
The main reason for this lack of studies is clear: the intricate presence of multiple isoforms makes this kind of approach a titanic enterprise. Although the prognostic value of TP53 is undeniable in certain populations of cancer patients (Olivier et al., 2009), the reason why its mutational status is not clinically relevant in other tumours is unclear. It is conceivable that harbouring a wild-type TP53 does not reflect the fitness of this tumour-suppressor pathway, since TP53 functions can be blocked by the TP73 and TP63 variants with oncogenic potential.
In this sense, it has been reported a strong correlation between the ΔTAp73 isoforms and the TP53 status in a cohort of 100 patients with ovarian cancer. Higher expression levels of ΔNp73 and ΔN’p73 isoforms were found in wild type TP53 tumors than in TP53 mutants, with no differences in the expression levels for TAp73. Moreover, the same authors showed that patients with TP53 mutations and a high expression of ΔTAp73 isoforms correlated with a failed chemotherapy response and shorter significant survival (Concin et al., 2005).
One year later, the same group analysed the prognostic value of TP73 expression profiles and TP53 mutational status in a cohort of 35 matched normal/tumoural samples from peritoneum or ovary and endometrium or ovary, respectively (Becker et al., 2006). The authors in that study developed a statistical model to analyse if the TP73 isoforms and TP53 expression levels could be combinatorial biomarkers for predicting outcome.
Although the TP73 variants expression was found to be upregulated in the tumoural tissue, no correlation was observed with the clinical outcome of the patients. However, the integration of the TP73 variants expression profiles and the TP53 status predicted correctly the outcome in 35% of patients. These reports provide interesting results for the TP53-TP73 cross-talk and the value of the combined alteration of both genes in the prognostic prediction of patients with gynaecological cancer.
In addition, it has been reported in breast cancer another interesting functional relationship for the TP53 family (Leong et al., 2007). The authors observed co-expression of ∆Np63α and TAp73 associated with cisplatin sensitivity but not with other chemotherapeutic drugs used for the treatment of breast cancer, in a subset of triple-negative breast tumours, which mainly display TP53 mutation. This response to cisplatin is the result of the physical dissociation-dependent phosphorylation of ABL1 of TAp73 from ΔNp63α and the subsequent transcription from the pro-apoptotic TAp73 effectors. Remarkably, this study significantly identified a subset of triple-negative breast cancer patients, who are generally resistant to conventional approaches to breast tumour therapy using cisplatin treatment.
Bahnassy and collaborators reported that disease-free survival of colon cancer patients correlated with the deregulated expressionof TP73 isoforms, while overall survival correlated significantly with aberrant levels of TP73 and TP63 regardless of TP53 status (Bahnassy et al., 2014).
In head and neck squamous cell cancers, it has been well documented that a high percentage of patients carries TP53 mutations. In addition to TP53 alterations, Faridoni-Laurens and collaborators reported in this type of tumor a high deregulation of the TP73 variant. Moreover, the few cases harboring a wild-type TP53 were consistently accompanied by deregulation of the ΔTAp73 isoform levels. Collectively, these findings suggest the general deregulation of at least one member of the TP53 family in this tumor type and its important role in the process of carcinogenesis (Faridoni-Laurens et al., 2008). In addition, as noted above, the authors reported a slight trend between the low expression of ΔNp73 levels and a better overall survival. However, the combined alteration of TP53 and TP73 forms did not influence survival. Although these data do not support their value as a prognostic marker, we need to keep in mind that the low number of cases included in this study and the high representation of TP53 mutations could have masked the final results. The latter is supported by the fact that the authors clearly stated that TP53, TP73 and TP63 altogether are key players in head and neck squamous epithelium cell cancer homeostasis and orchestrate different aspects of the differentiation process (Faridoni-Laurens et al., 2001).
On the other hand, little is known about the involvement of the TP53 family in extrahepatic biliary tract carcinoma. Hong et al. (2007) described the deregulation of TP63 and TP73 in this tumour type and the correlation of the combined overexpression of both with a shorter overall survival compared to the overexpression of TP63 or TP73. The data reported by this group again highlight the important role in the tumorigenic process of the interactions between the different members of the TP53 family.
In lung cancer has been described divergent results. Uramoto et al. (2004) reported the positive regulation of ΔNp73 as a prognostic marker in patients with lung cancer. However, the Lo Iacono’s group described the overexpression of the ΔEx2p73 and ΔEx2/3p73 variants, together with the down-regulation of ΔNp73 and ΔN’p73 isoforms in non-small cell lung cancer (NSCLC). However, taking into account the histological differences of the tumoral samples, these findings may importantly indicate a specific histological dysregulation of TP73 variants in lung tumorigenesis. It is noteworthy that these authors observed that higher levels of TAp63 and ΔN’p73 in normal tissue surrounding the tumoural tissue correlate with poor outcome, representing independent prognostic markers for overall survival (Lo Iacono et al., 2011). The accumulated evidence points to the events that occur in the cells of the tumor microenvironment as crucial for the progression of cancer. In this regard, deregulation of members of the TP53 family has not been evaluated so far, but should be explored as it could give extraordinary information for the understanding of the different roles of the members of the TP53 family and eventually could serve to manage cancer patients.
In conclusion, although several reports highlight the importance of TP53 family members individually in the development of different types of tumors (see “The role of TP73 in other tumors” section); only a few refer to their in vivo interrelationship and its importance in the outcome of cancer patients. Additional studies are needed to further explore the cross-talk between the TP53 family variants, as evidence increasingly points to the importance of their combined role in tumorigenesis.
Perspectives and Final Remarks
It is striking how few conclusive published studies have been performed trying to assess the implication of the aberrant expression of different TP73 isoforms on the prognosis of cancer patients. On the one hand, in few reports a complete follow-up of the patient cohorts is performed. In other reports, the recruitment rate of patients is significantly low. In spite of this, the here discussed studies highlight the high relevance of the TP73 variants in the tumorigenic process and the impact of its expression on the prognosis of cancer patients. Importantly, it is noteworthy that the absence of good antibodies against truncated NH2-terminal proteins may discourage researchers from embarking on this risky and time-consuming scientific venture. Although studies could be performed at mRNA levels, it is common to find paraffin-embedded tissue collections in hospitals and other institutions where routine immunohistochemistry procedures might be performed. Therefore, a major effort to be made by the TP73 scientific community should consist of the development of this type of material to help advance in the knowledge of the TP73 pathway.
It is also important to remark that the whole scenario involving the different TP53 family members has become even more complicated since a complex intricate network has been described in vitro. Consequently, scarce studies considering the status of all the different isoforms of more than one gene have been reported focusing on trying to determine the altered combinatorial contribution of the different variants of TP53, TP73, and TP63 in the prediction of the outcome of cancer patient’s (Concin et al., 2004).
Importantly, since it has been reported that certain downstream target genes of TP73 are able to predict the outcome of cancer patients better than ΔTAp73 (Soldevilla et al., 2011), an important field of study should consist of identifying specific target genes for the ΔTAp73 involved in the different tumorigenic processes. Indeed, these target genes might be responsible for the oncogenic role of the ΔTAp73 variants (Soldevilla et al., 2011).
Another open field of study consists of trying to evaluate the expression levels of the different TP73 variants in cancer patients through a non-invasive method. This idea relies on the observation that multiple variants of TP73 have been detected in the bloodstream material released by tumors in colorectal cancer patients and in patients with Wilms tumors (Soldevilla et al., 2014, Song et al. 2016). Undoubtedly, the possibility to measure the different TP73 variants by a blood test should be certainly closer to the clinical routine and might help clinicians with the management of cancer patients. However, it remains a fairly new field for TP73, which has yet to be implemented.
Finally, since the expression of suppressor or neutral forms could mask the final data, we want to remark that the altered expression of specific TP73 isoforms may probably predict the outcome of cancer patients more accurately than its total levels. Therefore, the simultaneous detection of the different TP73 isoforms not only could help in the management of cancer patients but also should definitely help deciphering the specific functions and involvement of each variant in the tumorigenesis process. Therefore, based on currently scientifically available data, TP73 isoforms are more and more emerging as attractive molecules to further investigate their use as prognostic markers and therapeutic targets of intervention in the clinical setting. Accordingly, inclusion in clinical trials of the expression levels of these forms along with the classical markers already included in the patient’s clinical records will help to show whether variants of TP73 could provide more accurate outcome predictions and get a better management and personalized treatments of cancer patients.
CONFLICT OF INTEREST
We declare no conflict of interest.
We acknowledge ISCIII, FIS, and FEDER for funding our current research through the PI15/00246 grant, and the grant from the MINECO (SAF2014-53209-R) and CIBERONC.
Amelio I, Inoue S, Markert EK, Levine AJ, Knight RA, Mak TW, et al. 2015. TAp73 opposes tumor angiogenesis by promoting hypoxia-inducible factor 1α degradation. Proc Natl Acad Sci U S A. 112, 226-31.
Bahnassy AA, Zekri AR, Salem SE, Abou-Bakr AA, Sakr MA, Abdel-Samiaa AG, et al. 2014. Differential expression of p53 family proteins in colorectal adenomas and carcinomas: Prognostic and predictive values. Histol Histopathol 29, 207-216.
Becker K, Pancoska P, Concin N, Vanden Heuvel K, Slade N, Fischer M, et al. 2006. Patterns of p73 N-terminal isoform expression and p53 status have prognostic value in gynecological cancers. Int J Oncol 29, 889–902.
Casciano I, Mazzocco K, Boni L, Pagnan G, Banelli B, Allemanni G, et al. 2002. Expression of DeltaNp73 is a molecular marker for adverse outcome in neuroblastoma patients. Cell Death Differ 9, 246–251.
Castellino RC, De Bortoli M, Lin LL, Skapura DG, Rajan JA, Adesina AM, et al. 2007. Overexpressed TP73 induces apoptosis in medulloblastoma. BMC Cancer 7, 127.
Chen YK, Huse SS, Lin LM. 2003. Differential expression of p53, p63 and p73 proteins in human buccal squamous-cell carcinomas. Clin Otolaryngol Allied Sci 28, 451–455.
Chen Y, Wang X, Li F, Zhang L, Ma L, Liu Y. 2015. Relationship between expression of P27, Fragile Histidine Triad (FHT), phosphatase and tensin homolog deleted on chromosome ten (PTEN), P73, and prognosis in esophageal squamous cell carcinoma. Ann Diagn Pathol. 19, 33-36.
Choi HR, Batsakis JG, Zhan F, Sturgis E, Luna MA, El-Naggar AK. 2002. Differential expression of p53 gene family members p63 and p73 in head and neck squamous tumorigenesis. Hum Pathol 33, 158–164.
Concin N, Becker K, Slade N, Erster S, Mueller-Holzner E, Ulmer H, et al. 2004. Transdominant DeltaTAp73 isoforms are frequently up-regulated in ovarian cancer. Evidence for their role as epigenetic p53 inhibitors in vivo. Cancer Res 64, 2449–2460.
Concin N, Hofstetter G, Berger A, Gehmacher A, Reimer D, Watrowski R, et al. 2005. Clinical relevance of dominant-negative p73 isoforms for responsiveness to chemotherapy and survival in ovarian cancer: evidence for a crucial p53-p73 cross-talk in vivo. Clin Cancer Res 11, 8372–8383.
De la Fuente M, Jones M-C, Santander-Ortega MJ, Mirenska A, Marimuthu P, Uchegbu I et al. 2015. A nano-enabled cancer-specific ITCH RNAi chemotherapy booster for pancreatic cancer. Nanomedicine 11, 369-377.
Domínguez G, García JM, Peña C, Silva J, García V, Martínez L, et al. 2006. DeltaTAp73 upregulation correlates with poor prognosis in human tumors: putative in vivo network involving p73 isoforms, p53, and E2F-1. J Clin Oncol 24, 805–815.
Faridoni-Laurens L, Bosq J, Janot F, Vayssade M, Le Bihan ML, Kaghad M, et al. 2001. P73 expression in basal layers of head and neck squamous epithelium: a role in differentiation and carcinogenesis in concert with p53 and p63? Oncogene 20, 5302–5312.
Faridoni-Laurens L, Tourpin S, Alsafadi S, Barrois M, Temam S, Janot F, et al. 2008. Involvement of N-terminally truncated variants of p73, deltaTAp73, in head and neck squamous cell cancer: a comparison with p53 mutations. Cell Cycle 7, 1587–1596.
Ferru A, Denis S, Guilhot J, Gibelin H, Tourani JM, Kraimps JL, et al. 2006. Expression of TAp73 and DeltaNp73 isoform transcripts in thyroid tumours. Eur J Surg Oncol 32, 228–230.
Fillippovich I, Sorokina N, Gatei M, Haupt Y, Hobson K, Moallem E, et al. 2001. Transactivation-deficient p73alpha (p73Deltaexon2) inhibits apoptosis and competes with p53. Oncogene 20, 514–522.
Ganzetti G, Rubini C, Campanati A, Zizzi A, Molinelli E, Rosa L, et al. 2015. IL-17, IL-23, and p73 expression in cutaneous melanoma: a pilot study. Melanoma Res. 25, 232-238.
Grob TJ, Novak U, Maisse C, Barcaroli D, Lüthi AU, Pirnia F, et al. 2001. Human delta Np73 regulates a dominant negative feedback loop for TAp73 and p53. Cell Death Differ 8, 1213–1223.
Guan M, Chen Y. 2005. Aberrant expression of DeltaNp73 in benign and malignant tumours of the prostate: correlation with Gleason score. J Clin Pathol 58, 1175–1179.
Hamdy MS, El-Saadany ZA, Makhlouf MM, Salama AI, Ibrahim NS, Gad AA.2016. TAp73 and ΔNp73 relative expression in Egyptian patients with lymphoid neoplasms. Tumori. 10.5301/tj.5000506
Hong SM, Cho H, Moskaluk CA, Yu E, Zaika AI. 2007. p63 and p73 expression in extrahepatic bile duct carcinoma and their clinical significance. J Mol Histol 38, 167–175.
Isakoff SJ, Mayer EL, He L, Traina TA, Carey LA, Krag KJ, et al. 2015. TBCRC009: A multicenter Phase II clinical trial of platinum monotherapy with biomarker assessment in metastatic triple-negative breast cancer. J Clin Oncol 33, 1902-1909.
Ishimoto O, Kawahara C, Enjo K, Obinata M, Nukiwa T, Ikawa S, 2002. Possible oncogenic potential of DeltaNp73: a newly identified isoform of human p73. Cancer Res. 62, 636–641.
Ito Y, Takeda T, Wakasa K, Tsujimoto M, Sakon M, Matsuura N. 2001. Expression of p73 and p63 proteins in pancreatic adenocarcinoma: p73 overexpression is inversely correlated with biological aggressiveness. Int J Mol Med 8, 67–71.
Ito Y, Uramoto H, Funa K, Yoshida H, Jikuzono T, Asahi S, et al. 2006. Delta Np73 expression in thyroid neoplasms originating from follicular cells. Pathology 38, 205–209.
Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. 2011. Global cancer statistics. CA Cancer J Clin 61, 69–90.
Jost CA, Marin MC, Kaelin WG Jr. 1997. p73 is a simian [correction of human] p53-related protein that can induce apoptosis. Nature 389, 191–194.
Kilic E, Brüggenwirth HT, Meier M, Naus NC, Beverloo HB, Meijerink JP, et al. 2008. Increased expression of p73Deltaex2 transcript in uveal melanoma with loss of chromosome 1p. Melanoma Res 18, 208–213.
Killick R, Niklison-Chirou M, Tomasini R, Bano D, Rufini A, Grespi F, et al. 2011. p73: a multifunctional protein in neurobiology. Mol Neurobiol 43, 139–146.
Leong CO, Vidnovic N, DeYoung MP, Sgroi D, Ellisen LW. 2007. The p63/p73 network mediates chemosensitivity to cisplatin in a biologically defined subset of primary breast cancers. J Clin Invest 117, 1370–1380.
Leupin N, Lüthi A, Novak U, Grob TJ, Hügli B, Graber H, et al. 2004. P73 status in B-cell chronic lymphocytic leukaemia. Leuk. Lymphoma 45, 1205–1207.
Lin YL, Sengupta S, Gurdziel K, Bell GW, Jacks T, Flores ER. 2009. p63 and p73 transcriptionally regulate genes involved in DNA repair. PLoS Genet. 5 e1000680.
Liu SS, Chan KYK, Cheung ANY, Liao XY, Leung TW, Ngan HYS. 2006. Expression of deltaNp73 and TAp73alpha independently associated with radiosensitivities and prognoses in cervical squamous cell carcinoma. Clin Cancer Res 12, 3922–3927.
Lo Iacono M, Monica V, Saviozzi S, Ceppi P, Bracco E, Papotti M, et al. 2011. p63 and p73 isoform expression in non-small cell lung cancer and corresponding morphological normal lung tissue. J Thorac Oncol 6, 473–481.
Lucena-Araujo AR, Kim HT, Thomé C, Jacomo RH, Melo RA, Bittencourt R, et al. 2015. High ΔNp73/TAp73 ratio is associated with poor prognosis in acute promyelocytic leukemia. Blood 126, 2302-2306.
Lunghi P, Costanzo A, Levrero M, Bonati A. 2004. Treatment with arsenic trioxide (ATO) and MEK1 inhibitor activates the p73-p53AIP1 apoptotic pathway in leukemia cells. Blood 104, 519–525.
Meier M, den Boer ML, Meijerink JPP, Broekhuis MJC, Passier MMCJ, van Wering ER, et al. 2006. Differential expression of p73 isoforms in relation to drug resistance in childhood T-lineage acute lymphoblastic leukaemia. Leukemia 20, 1377–1384.
Melino G, De Laurenzi V, Vousden KH. 2002. p73: Friend or foe in tumorigenesis. Nat Rev Cancer 2, 605–615.
Müller M, Schilling T, Sayan AE, Kairat A, Lorenz K, Schulze-Bergkamen H, et al. 2005. TAp73/Delta Np73 influences apoptotic response, chemosensitivity and prognosis in hepatocellular carcinoma. Cell Death Differ 12, 1564–1577.
O’Nions J, Brooks LA, Sullivan A, Bell A, Dunne B, Rozycka M, et al. 2001. p73 is over-expressed in vulval cancer principally as the Delta 2 isoform. Br J Cancer 85, 1551–1556.
Olivier M, Petitjean A, Marcel V, Pétré A, Mounawar M, Plymoth A, et al. 2009. Recent advances in p53 research: an interdisciplinary perspective. Cancer Gene Ther 16, 1–12.
Ott S, Geiser T. 2012. [Epidemiology of lung tumors]. Ther Umsch 69, 381–388.
Park HR, Kim YW, Park JH, Maeng YH, Nojima T, Hashimoto H, Park YK. 2004. Low expression of p63 and p73 in osteosarcoma. Tumori 90, 239–243.
Puig P, Capodieci P, Drobnjak M, Verbel D, Prives C, Cordon-Cardo C, et al. 2003. p73 Expression in human normal and tumor tissues: loss of p73alpha expression is associated with tumor progression in bladder cancer. Clin Cancer Res 9, 5642–5651.
Remke M, Hielscher T, Northcott PA, Witt H, Ryzhova M, Wittmann A, et al. 2011. Adult medulloblastoma comprises three major molecular variants. J Clin Oncol 29, 2717–2723.
Rizzo MG, Giombini E, Diverio D, Vignetti M, Sacchi A, Testa U, et al. 2004. Analysis of p73 expression pattern in acute myeloid leukemias: lack of DeltaN-p73 expression is a frequent feature of acute promyelocytic leukemia. Leukemia 18, 1804–1809.
Romani M, Scaruffi P, Casciano I, Mazzocco K, Lo Cunsolo C, Cavazzana A, et al. 1999. Stage-independent expression and genetic analysis of tp73 in neuroblastoma. Int J Cancer 84, 365–369.
Rufini A, Agostini M, Grespi F, Tomasini R, Sayan BS, Niklison-Chirou MV, et al. 2011. p73 in Cancer. Genes Cancer 2, 491–502.
Scian MJ, Carchman EH, Mohanraj L, Stagliano KER, Anderson MAE, Deb D, et al. 2008. Wild-type p53 and p73 negatively regulate expression of proliferation related genes. Oncogene 27, 2583–2593.
Schuster A, Schilling T, De Laurenzi V, Koch AF, Seitz S, Staib F, Teufel A, et al. 2010. ΔNp73β is oncogenic in hepatocellular carcinoma by blocking apoptosis signaling via death receptors and mitochondria. Cell Cycle. 9, 2629-2639.
Soldevilla B, Díaz R, Silva J, Campos-Martín Y, Muñoz C, García V, et al. 2011. Prognostic impact of ΔTAp73 isoform levels and their target genes in colon cancer patients. Clin Cancer Res 17, 6029–6039.
Soldevilla B, Rodríguez M, San Millán C, García V, Fernández-Periañez R, Gil-Calderón B, et al. 2014. Tumor-derived exosomes are enriched in ΔNp73, which promotes oncogenic potential in acceptor cells and correlates with patient survival. Hum Mol Genet. 23, 467-78.
Song D, Yue L, Wu G, Ma S, Yang H, Liu Q, et al. 2016. Evaluation of promoter hypomethylation and expression of p73 as a diagnostic and prognostic biomarker in Wilms’ tumour. J Clin Pathol., 69, 12-18.
Stiewe T, Pützer BM. 2000. Role of the p53-homologue p73 in E2F1-induced apoptosis. Nat Genet 26, 464–469.
Stiewe T, Theseling CC, Pützer BM. 2002a. Transactivation-deficient Delta TA-p73 inhibits p53 by direct competition for DNA binding: implications for tumorigenesis. J Biol Chem 277, 14177–14185.
Sun XF. 2002. p73 overexpression is a prognostic factor in patients with colorectal adenocarcinoma. Clin Cancer Res 8, 165–170.
Takahashi H, Ichimiya S, Nimura Y, Watanabe M, Furusato M, Wakui S, et al. 1998. Mutation, allelotyping, and transcription analyses of the p73 gene in prostatic carcinoma. Cancer Res 58, 2076–2077.
Thakur AK, Nigri J, Lac S, Leca J, Bressy C, Berthezene P, et al. 2016. TAp73 loss favors Smad-independent TGF-β signaling that drives EMT in pancreatic ductal adenocarcinoma. Cell Death Differ 23: 1358-1370.
Tomasini R, Tsuchihara K, Wilhelm M, Fujitani M, Rufini A, Cheung CC, et al. 2008. TAp73 knockout shows genomic instability with infertility and tumor suppressor functions. Genes Dev 22, 2677–2691.
Toumi AA, El Hadj OEA, Ben Mahmoud LK, Ben Hmida AEM, Chaar I, Gharbi L, et al. 2010. The prognostic value of p73 overexpression in colorectal carcinoma: a clinicopathologic, immunohistochemical, and statistical study of 204 patients. Appl Immunohistochem Mol Morphol 18, 128–136.
Tuve S, Wagner SN, Schittek B, Pützer BM. 2004. Alterations of DeltaTA-p 73 splice transcripts during melanoma development and progression. Int J Cancer 108, 162–166.
Uramoto H, Sugio K, Oyama T, Nakata S, Ono K, Morita M, et al. 2004. Expression of deltaNp73 predicts poor prognosis in lung cancer. Clin Cancer Res 10, 6905–6911.
Vilgelm A, Wei JX, Piazuelo MB, Washington MK, Prassolov V, El-Rifai W, et al. 2008. DeltaNp73alpha regulates MDR1 expression by inhibiting p53 function. Oncogene 27, 2170–2176.
Vilgelm AE, Hong SM, Washington MK, Wei J, Chen H, El-Rifai W, et al. 2010. Characterization of ΔNp73 expression and regulation in gastric and esophageal tumors. Oncogene 29, 5861–5868.
Wager M, Guilhot J, Blanc JL., Ferrand S, Milin S, Bataille B, et al. 2006. Prognostic value of increase in transcript levels of Tp73 DeltaEx2-3 isoforms in low-grade glioma patients. Br J Cancer 95, 1062–1069.
Weber A, Bellmann U, Bootz F, Wittekind C, Tannapfel A. 2002. Expression of p53 and its homologues in primary and recurrent squamous cell carcinomas of the head and neck. Int J Cancer 99, 22–28.
Wilhelm MT, Rufini A, Wetzel MK, Tsuchihara K, Inoue S, Tomasini R, et al. 2010. Isoform-specific p73 knockout mice reveal a novel role for delta Np73 in the DNA damage response pathway. Genes Dev 24, 549–560.
Yang A, Walker N, Bronson R, Kaghad M, Oosterwegel M, Bonnin J, et al. 2000. p73-deficient mice have neurological, pheromonal and inflammatory defects but lack spontaneous tumours. Nature 404, 99–103.
Yokomizo A, Mai M, Bostwick DG, Tindall DJ, Qian J, Cheng L, et al. 1999. Mutation and expression analysis of the p73 gene in prostate cancer. Prostate 39, 94–100.
Yong M, Yang L, Suyila Q, Han W, Yuan H, Zhao C, et al. 2014. Expression and clinical implications of P53, P63, and P73 protein in malignant tumor of the parotid gland. Turk J Med Sci. 44, 875-882.
Zaika AI, Slade N, Erster SH, Sansome C, Joseph TW, Pearl M, et al. 2002. DeltaNp73, a dominant-negative inhibitor of wild-type p53 and TAp73, is up-regulated in human tumors. J. Exp. Med. 196, 765–780.
Zitterbart K, Zavrelova I, Kadlecova J, Spesna R, Kratochvilova A, Pavelka Z, et al. 2007. p73 expression in medulloblastoma: TAp73/DeltaNp73 transcript detection and possible association of p73alpha/DeltaNp73 immunoreactivity with survival. Acta Neuropathol 114, 641–650.
Zhao W, Zhang B, Guo X, Zhang X, Hu J, Hu X, et al. 2014. Expression of Ki-67, Bax and p73 in patients with hilar cholangiocarcinoma. Cancer Biomark 14, 197-202.
Zhu X, Pan X, Yang Z, Xing P, Zhang Y, Li F, et al. 2015. Experssion and prognostic significance of TAp73 and ∆Np73 in FIGO stage I-II cervical squamous cell carcinoma. Oncol Lett 9, 2090-2094