Mechanism of acquired resistance in the treatment of ccRCC

CLEAR CELL RENAL CELL CARCINOMA (ccRCC)

Histopathology and frequency

Renal carcinoma are separated in two groups: 1) “Non- papillary renal carcinoma” have a loss of heterozygosity in the chromosome 3 containing the VHL gene, due to somatic mutation (in 50% of the cases). In 10 % of the cases there is a inactivation of the VHL gene by epigenetic changes as hypermethylation in its promoter); 2) those named “Papillary Renal Carcinoma” that are characterized with chromosomal abnormalities at loci other than chromosome 3. Genetically, these tumors are characterized by trisomies (chromosomes 3q, 7, 12, 16, 17, and 20) and loss of the Y chromosome (Störkel 2000) (Zar 1994). According to the World Health Organization , from all subtypes of RCC, non-papillary ccRCC is diagnosed in the 70-85% of cases versus only 10-15% papillary RCC and 4-5% chromophobe RCC.
Non-papillary clear cell of Renal carcinoma is considered to belong to the malignant group of kidney neoplasm. In histology this renal cortical tumor is typically characterized by epithelial cells with clear cytoplasm, because it is filled with lipids and glycogen that are dissolved during the histologic processing, creating the distinct clear cytoplasm (Haddad 1993). It is frequent as well, to find foci in which cells have an eosinophilic staining, often in the case of high grade tumors and adjacent to areas with necrosis or haemorrhage. These tumors cells have an acinar or very compact –alveolar growth (Storkel 2000).

Somatic Genetics

As previously described, the majority of ccRCC does not have an origin in the von Hippel Lindau disease. Nonetheless, deletions in the short arm of the chromosome 3 leading to partial or total inactivation of the VHL gene has shown to be a constant in the sporadic clear cell Renal Carcinoma (Brouch 2010) (Foster 1994) (Gnarra 1994). In the von Hippel Lindau disease, mutation in the tumor suppressor gene is inherited and the malignant kidney cancer arises due to a second sporadic mutation in the wild type allele gene. In the case of sporadic ccRCC, the mutations occur on the same cell type after somatic inactivation of both alleles of the VHL gene by allelic deletion, mutation, or epigenetic silencing in 70% or more of the cases. (Gnarra 1994) (Knudson 1985) (Herman 1994) (Latif 1993) (Schram 2002). These data suggest that the VHL gene is most likely a tumor suppressor gene in sporadic clear cell RCC.

Kidney cancer stages, treatment and prognosis

At first basis, the type of treatment that will be applied to a patient with RCC is determined by the probability of cure, that in most cases is directly related on how early the patient is diagnosed, the state of the tumor and the degree of tumor dissemination (American Cancer Facts 2015; Sachdeva 2016; RCC is mostly a silenced disease, probably because kidneys are localized deep in the body and there may not be any symptoms until the tumor size is large, only 10% of patients will approach their doctor due to classic triad of flank pain, hematuria, and flank mass.
Chart N°1 reveals the different sizes and stages of RCC. Chart N°2 describes the 5 years survival prognosis. More than 50% of patients with early stage renal cell carcinoma are cured, but according to the Cancer Center treatments of America and the Surveillance, Epidemiology and End Results (SEER) database of the National Cancer Institute in USA, the 5 years survival outcome for stage IV disease is poor (around 10% of patients).

TREATMENT OF METASTATIC RCC Surgery

Partial nephrectomy has been so far the first line of treatment for organ-confined ccRCC (tumors smaller than 4 cm), and in some cases a total nephrectomy can be applied in larger tumors. In the last years there have been an increased interest in including studies that will evaluate not only the appearance of local recurrence after surgery but also the way metastasis respond to drug therapies in presence or absence of the primary tumor (Coppin 2011; Lams 2006).

Adjuvants therapies

Adjuvants therapies are the treatments that are given in addition to the primary or initial therapy to maximize their effectiveness. Usually, in different types of cancer, the adjuvant is the treatment given after surgery where the detectable disease has been removed but when there is still risk ofrelapse due to non-detectable residual disease (Pal 2014; National Cancer Institute: Dictionary of cancer terms).In this section I will discuss the therapies that are used in a monotherapy or as an adjuvant in the systemic treatment of mRCC.

Immunotherapies

Interferon α (IFN α)

Interferon α was the first cytokine therapy approved for the systemic treatment of mRCC.
This natural glycoprotein expressed by leukocytes, stimulates natural killer cells (NK), decreases cell proliferation through the inhibition of cyclin kinases, increases immunogenicity of tumor cells and inhibits angiogenesis (Fossa 2001; Goldstein 1988; Coppin 2000; Oliver 1989).

Targeted Therapies

Overall, the systemic therapy of advanced RCC has not been satisfactory since, as discussed before, cytokine treatments have shown limited efficacy as well as high toxicity. Since then, the better understanding of the signaling pathways involved in the development of ccRCC has led to the development of novel therapies that target specifically the dysregulated pathways inside the tumor cell and/or the cells that are in the tumor environment.
As described before, the inactivation of the VHL gene leads to the accumulation of the transcription factor HIF, over-expression of vascular endothelial growth factor (VEGF) and platelet-derived growth factor (PDGF) (Wiessener 2001; Cookman 2000; lliopoulos 1996).
These proteins are known to promote angiogenesis, a key process in tumor development and progression (Takahashi 1994).
As well, the mammalian target of rapamycin (mTOR), a protein kinase, is in the upstream pathway regulating not only HIF but also other factors as angiogenesis, cell proliferation, metabolism that are critical for the pathogenesis of mRCC (Hudson 2002).

mTOR inhibitors

Temsilorimus

Temsilorimus is an inhibitor of the mammalian target rapamycin mTOR. Treatment of cancer cells with this agent induces cell cycle arrest and inhibition of angiogenesis through regulation of VEGF that is under the regulation of the HIF pathway (Hudson 2002; Faivre 2006). Its efficacy was investigated in a randomized phase-III study (Hudes 2007) that included patients that had poor prognosis factors according to the Memorial Sloan-Kettering Cancer Center (MSKCC) prognostic criteria (Motzer 2002). The comparison was done between three arms: The first one, treatment with only IFN α; the second one, treatment with Temsilorimus and the third one, the combination of both. The results of ORR did not differ significantly between arms: INFα (4.8%), Temsirolimus (8.6%), and combination therapy (8.1%). PFS showed a benefit of 2 months for the second group (Temsilorimus alone, compared to the first group of INF-α alone). OS was higher as well in the second group (10.9 versus 7.3 months). However, when patients on the combination regimen were compared with the interferon-α group, PFS and OS were similar, being 4.7 and 8.4 months, respectively. Nevertheless, there are high toxicities associated to this treatment. In 11% of the patients the most common grade 3 and 4 side effects are asthenia and hyperglycemia. Based on positive survival and PFS results,Temsirolimus is recognised in recent guidelines as a first-line treatment option for patients withmRCC who have poor MSKCC prognostic factors.

Everolimus

Compared to Temsilorimus, this mTOR inhibitor is administrated orally and has a different active form. In RECORD-1 trial (Motzer 2008), patients with metastatic renal cell carcinoma, which had progressed on Sunitinib, Sorafenib, or both, were randomly assigned into a group treated with Everolimus versus a group treated with placebo. The outcome of this study showed no difference in the median OS between groups at the end of double-blind analysis. PFS was evaluated in an unblinded analysis, since at disease progression patients with placebo were offered to turn into Everolimus. Median PFS was significantly improved on this late group compared to the placebo group (4.9 versus 1.9 months). Lack of increased OS was likely the result of the crossover design of the study (Motzer 2010).
Toxicities associated to this treatment are mostly of grade 3 and are observed in 15% of patients. This toxicities, stomatitis (44%), infections (37%), asthenia (33%), fatigue (31%), diarrhea (30%), rash (29%), nausea (26%), anorexia (25%), peripheral edema (25%), vomiting (24%), thrombocytopenia (23%), hypercholesterolemia (77%), hypertriglyceridemia (73%), and hyperglycemia (57%) are comparable to those induced by the Temsilorimus treatment (Hudson 2007). In addition, Everolimus-induced pneumonitis is a side effect in 4% of patients (Motzer 2010).
Based on the data from this trial, Everolimus is now the recommended therapy in patients who have progressed on prior VEGF-targeted therapy.

MET inhibitors

Cabozantinib

Cabozantinib has arise as a novel alternative and second line treatment for patients that have failed to VEGFR therapy. This inhibitor of tyrosine kinases including MET, VEGFR, and AXL has shown to improve the median overall survival (21,4 months) compared to everolimus (16,5 months).
Cabozantinib treatment also resulted in improved progression-free survival (7.4 months in patients treated with cabozantinib compared to 3,8 months in patients treated with everolimus ) . The most frequent type of adverse effects were grade 3 or 4. hypertension (49 [15%] in the cabozantinib group vs 12 [4%] in the everolimus group), diarrhoea (43 [13%] vs 7 [2%]), fatigue (36 [11%] vs 24 [7%]), palmar-plantar erythrodysaesthesia syndrome (27 [8%] vs 3 [1%]), anaemia (19 [6%] vs 53 [17%]) , between others. (Choueiri 2016, Grassi 2016)
Based on these results, cabozantinib has been approved for the treatment of mRCC as a second line therapy.

HYPOXIA INDUCIBLE FACTORS

Structures and tissue expression level

The HIF proteins are part of a family with 3 members (HIF-1α, HIF-2α and HIF-3α) and two HIFβ members (HIF-1β and HIF-2β). HIF-1β is also called Aryl hydrocarbon receptor nuclear translocator (ARNT). Both HIF-1α and HIF-2α have two transcriptional activation domains, the N-terminal transactivation domain (NTAD) and the C- terminal activation domain (CTDA) that activates transcription when bound to the DNA (Kaelin 2008).
As discussed before, the degradation of the HIF-1α and HIF-2α isoforms is dependent on their recognition by pVHL of one or two hydroxylated prolyl residues within their NTAD,generated by PHD. Oxygen dependent post-translational modifications will lead to ubiquitination and later degradation through the proteasome (Ivan 2001; Jaakkola 2001; Ju 2001). This is described in the next figure N°10.

NOVELS TARGETS AGAINST ccRCC

ATAXIA TELANGESTASIA MUTATED KINASE (ATM)

As it will be later discuss, two important kinases that where never described before as potentials therapeautic targets against ccRCC, came out from our screening. In the next chapter I will describe the main characteristic and properties of each kinase. As well, the inhibitors that have been develop to inhibit these kinases will be discussed.

ATM structure

The ATM protein is a very large protein of 370 KDa that is encoded in the human chromosome 44q22-23 and shares high homology with the mouse protein (84%). Homologsof this protein have been found in all eukaryotes.

ATM cell function

ATM role in DNA damage response activity.

Double strand breaks (DSBs) can be a direct consequence of DNA damage induce by X-rays, ionising radiation (IR), or radio-mimetic chemicals or can be produce indirectly, for example, by reactive oxygen species (ROS) that are often generated as by-products of normal cellular metabolic processes.
DSBs can also be generated during replication fork collapse, which occurs after prolonged replication fork stalling or when the replication machinery encounters certain lesions such as DNA single strand breaks (SSBs) or an abortive topoisomerase complexe (reviewed in Shiloh 2003).
Upon the formation of a DSB, the first factor to sense the damage is the MRE11/RAD50/NBS1 (MRN) complex stabilizing the free DNA ends (Williams 2007). Once bound to a DSB, the MRN complex triggers the activation of ATM via its association with the C-terminus of NBS1 (Falck 2005). In the absence of DNA damage, ATM exists as an inactive dimer. Following break recognition by the MRN complex the inactive ATM dimers dissociates and forms active monomers. This damage-induced event occurs by autophosphorylation of ATM in trans on serine (S) 1981, S367, S1893 and S2996, as well as acetylation on lysine 3016 by the histone acetyltransferase Tip60 (Bakkenist 2003, Kozlov 2011).
One of the first targets of the activated ATM is the histone H2A variant (γH2AX), which becomes phosphorylated on S139 (Rogakou 1998).
The phosphorylation and dephosphorylation of γH2AX leads to the recruitment of the mediator protein called mediator of DNA damage checkpoint 1 (MDC1), which interacts directly with the γH2AX, via its BRCA1 C-terminal (BRCT) domain (Lee 2005).
The binding of γH2AX to MDC1 stimulates the recruitment of additional MRN complexes to the site of DNA damage, which serves to recruit other activated ATM therefore spreading the γH2AX signal along the chromatin in either side of the lesion. Using this positive ATMdependent feedback loop, this process results in amplification of the intracellular signal generated by the DNA damage (Chapman 2008, Melander 2008, Spycher 2008). MDC1 has also the function to recruit others DNA damage response (DDR) proteins to the site of damage. Through its interaction with ATM, the N-terminus of MDC1 becomes phosphorylated, creating a binding plateform recognized by the forkhead-associated (FHA) domain of the E3 ubiquitin ligase, RNF8 (Huen 2007, Kolas 2007). RNF8 catalyses the ubiquitylation of H2A and γH2AX, surrounding the break. The ubiquitylation of H2A-type histones signals the recruitment of another E3-ubiquitin ligase, RNF168. RNF168, together with the E2 conjugating enzyme UBC13, catalyses the lysine 63 (K63) -linked poly-ubiquitylation of H2A/γH2AX, which is thought to facilitate chromatin relaxation proximal to the break and also create an interaction plateform for the recruitment of downstream DNA repair/checkpoint proteins, such as BRCA1 and 53BP1 (Doil 2009, Stewart 2009).

ATM Dependent Modulation of Signalling Pathways Outside DDR Implicated in Cancer

ATM has been shown to be a key player in the maintenance of the genome stability and DNA damage response and that dysregulation of these pathways is strongly linked to cancer initiation and progression. Interestingly, ATM has also been found to have extended functions in signalling networks that sustain tumorigenesis, including oxidative stress, hypoxia, receptor tyrosine kinase and AKT serine-threonine kinase activation (Reviewed by Stagni 2014).
Alternative functions for ATM where found from a global proteome analysis performed by (Matsuoka 2007), who identified about 900 ATM targets in response to ionizing radiation (IR), among which several are proteins that exert important functions outside of the DNA damage response. Importantly, albeit ATM was originally identified to be mostly present in the nucleus, several reports have demonstrated also its cytoplasmic localization (Li 2009) and more recently to the mitochondria (Valentin-Vega 2012) and to participate to selective autophagy and to the peroxisomes homeostasis (Tripathi 2016), pointing the possibility that ATM participates in several signalling cascades others than DNA damage response. Some of these functions are summarized in the next table (TableN°7).

Modulation of ATM Activity in Cancer Therapy

ATM kinase activation and therefore DNA damage response can enable tumors to resist classical treatments as IR or other chemotherapeutic drugs. This response is beneficial for the tumor cells only when the tumor suppressor activity of DDR has been disabled, for example, by deactivating p53 through mutation or loss of expression (Halazonetis 2008).
As previously described, A-T-patients have increased radiosensitivity to IR. This observation raised the question of how to sensitize tumors to IR or other chemotherapeutic agents through the inhibition of ATM to overcome resistance (Jiang 2009, Bouwman 2012). So far, four ATM kinase inhibitors have been produced: (1) KU-55933 (Hickson 2004); (2) CP466722 (Rainey 2008) ; (3) KU-59403(Batey 2013) and (4) KU-60019 (Golding 2009). All these compound structures are represented in the figure N°14:

PROTEIN KINASE CK2

CK2 Structure

The protein kinase CK2, wrongly called casein kinase 2, was discovered almost 60 years ago from studies using casein as substrate (Burnett and Kennedy, 1954). It has been extensively studied showing that CK2 is a multifunctional, ubiquitously expressed and constitutively active, protein kinase that preferentially phosphorylates serine, threonine, and tyrosine residues within clusters of acidic residues (Pinna 1990; Basnet 2014).
The human enzyme structure mostly appears as a heterotetrameric protein complex with two catalytic subunits CK2α (42 kDa) and/or α’ (38 kDa) and one dimer of a regulatory subunit CK2β (28 kDa) (Allende 1995) (Guerra 1999) (Litchfield 2001).The α and α’ subunits exhibit 90% of homology in their catalytic domains as well they show similar action in their enzymatic activity in vitro (Figure N°15) (Bodenbach 1994, Litchfield 1990). CK2β represents the regulatory subunit, being at the centre of the tetrameric complex as a dimer, and with the two catalytic subunits attaching separately to this dimer (Boldyreff 1996, Chantalat 1999, Gietz 1995, Graham and Litchfield, 2000).
Although this subunit is not required for the activity of the catalytic subunits, CK2β binding regulates a large range of substrates that usually are not phosphorylated or only weakly phosphorylated. In opposite, only a few substrates are phosphorylated by the catalyticsubunit alone and not by the holoenzyme (Meggio 1992, Pinna 2002).
In addition, it has been reported that close association between CK2 and some of its substrates is often bridged by the CK2β dimer (Guerra 1999, Filhol 1992 ,Golden 2015, Deshiere 2013, Escarqueil 2000, Maizel 2002, Huillard 2010). This means that any change in CK2β expression might lead to a shift in the balance of phosphorylated CK2α- and holoenzyme-specific substrates.

Table des matières
INTRODUCTION
CHAPTER 1: Kidney and Cancer
1. WHY IS YOUR KIDNEY SO IMPORTANT?
Kidney Physiology
1.2 Kidney Functions
2. KIDNEY CANCER
Frequency and mortality
2.2 Causes of kidney cancer
A) Non genetic associated factors
B) Genetic associated risk factors
3. CLEAR CELL RENAL CELL CARCINOMA (ccRCC)
Histopathology and frequency
3.2 Somatic Genetics
3.3 Kidney cancer stages, treatment and prognosis
4. TREATMENT OF METASTATIC RCC
Surgery
4.2 Adjuvants therapies
• Immunotherapies
A) Interferon α (IFN α)
B) Interleukin-2 (IL-2)
• Targeted Therapies
A) VEGF/VEGF Receptor (VEGFR) inhibitors
B) mTOR inhibitors
C) MET inhibitors
Sequential and Combinational therapy
5.MECHANISM OF ACQUIRED RESISTANCE IN THE TREATMENT OF ccRCC
Resistance to Tyrosine Kinase Inhibitors
5.2 Resistance to mTOR inhibitors
CHAPTER 2: THE ROLE OF HYPOXIA IN ccRCC 
1.VON HIPPEL LINDAU TUMOR SUPPRESSOR GENE (VHL)
VHL gene products
1.2 VHL and the HIF axis
2. HYPOXIA INDUCIBLE FACTORS
Structures and tissue expression level
2.2 Implications of HIF-1α and HIF-2α in the ccRCC
2.3 HIF isoforms as prognosis factors
CHAPTER 3: NOVELS TARGETS AGAINST ccRCC
1. ATAXIA TELANGESTASIA MUTATED KINASE (ATM)
ATM structure
1.2 ATM cell function
A) ATM role in DNA damage response activity
B) ATM signaling in cancer cells
C) ATM Dependent Modulation of Signalling Pathways Outside DDR Implicated in Cancer
1.3 Modulation of ATM Activity in Cancer Therapy
2.PROTEIN KINASE CK2  CK2 Structure
2.2 Cellular functions in Cancer Disease.
2.3 CK2 Protein expression in Renal Cancer
2.4 CK2 and the HIF axis
2.5 CK2 is a druggable target
A) CK2 inhibitors
2.6 The CX-4945 inhibitor
A) Specificity and cell biology function
B) Clinical Phase trials
C) CX-4945 in combinational therapy
RESULTS
Search for novel therapeutic targets against ccRCC by High-Throughput screening
Set up of conditions for High-throughput Screening (HTS)
A) 786-O as cell line model
B) Kinases as targets
C) Cells seeding and Cell Viability Markers
D) Determination of Z’Score Factor
1.2 The Screening
1.3 Validation of hits through molecular biology studies
Discussion
Conclusions
RESULTS
PATENT: “Novel therapeutic drug combination against Kidney Cancer »
Introduction
Materials and Methods
Results
Claims
Tables and Figures
Figures legends
Discussion
SUPPLEMENTARY RESULTS
Study of the phenotype given by drug inhibitors
Intracellular vacuolar structures
Origin of vacuoles
A) Endocytosis
B) Early and Late endosomes
Mechanism of resistance
D) Lysosomal sequestration
E) Autophagy
Discussion
Conclusion
SUPPLEMENTARY RESULTS
Preliminary results on the mechanistic of CK2/ATM drug combination
ATM/CK2 dysregulation in ccRCC human samples
HIF-2α: The Redemption
HIF2α is a substrate for CK2
Discussion
FINAL CONCLUSION AND PERSPECTIVES
REFERENCES
APPENDIX 

Lire le rapport complet