The cell cycle is one of the most important regulatory mechanisms of cellular growth and proliferation. Dysregulation of this pathway is thought to be the first step in carcinogenesis of renal cell carcinoma (RCC), important for tumor invasion and metastases. Multiple different parts and regulators of the cell cycle are detectable via assays like immunohistochemistry and have been studied extensively. This review aims to provide an overview of current data regarding individual cell cycle and proliferative markers and their association with clinicopathologic parameters and their impact on prognosis for patients with RCC. Furthermore, the value of marker combinations is discussed.
Key points
- •
Biomarkers can provide additional information to routinely assess clinicopathologic features regarding prognosis of patients with clear cell renal cell carcinoma (ccRCC).
- •
Of the cell cycle and proliferative biomarkers, p53 and Ki-67 are the most studied markers and provide independent prognostic information in patients with ccRCC.
- •
Marker combinations of one or multiple pathways are thought to be superior to single markers due to complexity of carcinogenesis.
- •
Because of a lack of proper validation, no cell cycle or proliferative biomarkers are currently used in routine care to guide treatment decisions.
Introduction
Clear cell renal cell carcinoma (ccRCC) is the most common malignancy of the kidneys and comprises most of the estimated 61,560 cases of renal malignancies and 14,080 deaths in the United States in 2015. With increased availability of cross-sectional imaging, more renal masses are found incidentally before the appearance of systemic symptoms. The proportionate increase of small renal masses is tremendous from about 3% to 17% in the 1970s to up to 48% to 66% in current studies. Although around 20% to 30% of patients with ccRCC are initially metastatic, 20% to 40% of patients will relapse after initial curatively intended surgery. Unfortunately, patients who relapse often require systemic therapy and have a poor prognosis.
Improved prognostic classification apart from conventional TNM staging could provide a substantial increase in information, potentially with therapeutic implications. First, a personalized surveillance protocol could be established for each patient based on their risk, such that relapses could be detected early, making salvage metastasectomy a treatment possibility to provide long-term cure. Second, improved risk stratification may be important in design of trials using adjuvant therapies for patients with advanced disease. Furthermore, tissue biomarkers can be useful to select systemic therapies in patients with advanced disease in the presurgical setting to predict response to therapy more accurately. After assessment of biomarker profiles through biopsies, treatment choices might be made in these patients with information from biomarkers in addition to conventional histology. Last, with new targets like the recently introduced CTLA-4, PD-1, and PD-L1 inhibitors, a whole new class of drugs is available for which biomarkers are very important, and still missing, to distinguish possible responders from nonresponders as early as possible, preferably in advance. Therefore, prognostic as well as predictive biomarkers are urgently needed for patients with ccRCC.
Introduction
Clear cell renal cell carcinoma (ccRCC) is the most common malignancy of the kidneys and comprises most of the estimated 61,560 cases of renal malignancies and 14,080 deaths in the United States in 2015. With increased availability of cross-sectional imaging, more renal masses are found incidentally before the appearance of systemic symptoms. The proportionate increase of small renal masses is tremendous from about 3% to 17% in the 1970s to up to 48% to 66% in current studies. Although around 20% to 30% of patients with ccRCC are initially metastatic, 20% to 40% of patients will relapse after initial curatively intended surgery. Unfortunately, patients who relapse often require systemic therapy and have a poor prognosis.
Improved prognostic classification apart from conventional TNM staging could provide a substantial increase in information, potentially with therapeutic implications. First, a personalized surveillance protocol could be established for each patient based on their risk, such that relapses could be detected early, making salvage metastasectomy a treatment possibility to provide long-term cure. Second, improved risk stratification may be important in design of trials using adjuvant therapies for patients with advanced disease. Furthermore, tissue biomarkers can be useful to select systemic therapies in patients with advanced disease in the presurgical setting to predict response to therapy more accurately. After assessment of biomarker profiles through biopsies, treatment choices might be made in these patients with information from biomarkers in addition to conventional histology. Last, with new targets like the recently introduced CTLA-4, PD-1, and PD-L1 inhibitors, a whole new class of drugs is available for which biomarkers are very important, and still missing, to distinguish possible responders from nonresponders as early as possible, preferably in advance. Therefore, prognostic as well as predictive biomarkers are urgently needed for patients with ccRCC.
Cell cycle and cell proliferation
The cell cycle is one of the most important regulatory mechanisms of the human body because it controls rate of cell division and proliferation. The cell cycle is subdivided into different phases (G1, S, G2, and G0), which have to be completed in a certain order before cell division is completed. Control mechanisms are found in protein complexes consisting of cyclins and cyclin-dependent kinases. These complexes control orderly progression through the cell cycle. Progression of the cell cycle is achieved by phosphorylation of key components and subsequent release of inhibition at certain checkpoints. These processes are completed thousands of times each day in a human body. Loss of cell cycle regulation is thought to be the first step in carcinogenesis and an important contributor to tumor invasion as well as development of metastases.
The current article focuses on the prognostic role of cell cycle and proliferative markers in patients with ccRCC.
Prognostic value of cell cycle and proliferative markers
p53 and TP53
p53 is one of the major regulatory proteins in cell division and is often called the guardian of the cell cycle. It acts as a tumor suppressor by inhibition of cell cycle progression and induction of apoptosis in cells that suffer DNA damage. Dysfunctional p53 leads to loss of control of cell division and lack of apoptotic signals in affected cells. Mutations of TP53, the gene coding for p53, leads to extended half-life of p53 and accumulation of the protein in the nucleus, making it detectable by immunohistochemistry. However, immunohistochemistry is not able to differentiate wild-type p53 versus mutant p53. TP53 mutations are among the most frequent in human cancers, found in up to 50% of cases. The importance of the tumor-suppressive properties of p53 and the impact of p53 mutations becomes apparent when looking at the frequency of cancers in patients with germline mutations of p53 as in patients with Li-Fraumeni syndrome, who develop a diverse set of malignancies, including breast carcinomas, sarcomas, and brain tumors.
In ccRCC, many investigators have evaluated the role of p53 for individual tumor characteristics as well as the prognosis for oncologic outcomes. However, for the interpretation of results, it is important to consider possible pitfalls. First, the issue of cutoffs used and its implication on rate of expression must be discussed. Another issue to consider is the method of detection of p53. Most of the studies used immunohistochemistry for p53 evaluation, which harbors a large possibility for variable results. Depending on antibody (mostly DO-7) and cutoff of positivity (>1% to >20%) used in the different trials, results may differ significantly. As there are no regulations of cutoffs for p53 staining, the available studies are not always comparable. Also, validation studies are currently missing because suggested cutoffs were often developed in the datasets themselves. Nonetheless, there are still patterns that emerge regarding the significance of p53 in renal cell carcinoma.
Studies comparing primary and metastatic tumor sites found that p53 overexpression is seen more often in metastatic tissue (50%–85%) than in primary tumors (20%–35%), suggesting accumulation of mutations and dysregulation promoting aggressiveness along the course of disease. Overall, p53 overexpression seems to be lower in ccRCC (11.9%) in comparison to other histologic subtypes (27%–70%). However, it is important to emphasize that the altered expression rate is directly affected by the cutoff used to define alterations. The average range of p53 overexpression is suggested to be between 10% and 40% ( Table 1 ). Also, p53 accumulation is heterogeneous across tumor sections, further complicating interpretation.
| Study | No. of RCC (% ccRCC) | Cutoff for Alteration, % | Alteration Rate, % | Prognostic Value in UVA | Prognostic Value in MVA | Endpoint with Prognostic Value |
|---|---|---|---|---|---|---|
| Weber et al, 2014 | 145 (100) | — a | — | ✓ | ✗ | ↓ b DSS |
| Weber et al, 2013 | 132 (100) | >15 | 50.8 | ✗ | ✗ | — |
| Noon et al, 2012 | 97 (90) | >10 | 15.6 | ✗ | ✗ | — |
| Baytekin et al, 2011 | 104 (63.5) | >10 | 13.5 | ✗ | ✗ | — |
| Dahinden et al, 2010 | 527 (100) | NR | NR | ✗ | ✗ | — |
| Zubac et al, 2009 | 160 (100) | >10 | 53 | ✓ | ✓ | ↓ DSS |
| Sakai et al, 2009 | 153 (86.3) | >20 | 33.3 | ✗ | ✗ | — |
| Klatte et al, 2009 | 170 (100) | Any positive | NR | ✓ | ✓ | ↓ RFS |
| Perret et al, 2008 | 50 (0) | >20 | 48 | ✓ | ✗ | ↓ OS |
| Phuoc et al, 2007 | 119 (100) | >10 | 54 | ✓ | ✓ | ↓ DSS |
| Kankuri et al, 2006 | 117 (86) | >10 | 12.8 | ✓ | ✗ | ↓ b DSS |
| Kramer et al, 2005 | 117 (89) | >5 | 13.6 | ✗ | ✗ | — |
| Cho et al, 2005 | 92 (100) | >10 | 12 | ✓ | ✓ | ↓ DSS |
| Shvarts et al, 2005 | 193 (85) | >20 | 7.3 | ✓ | ✓ | ↓ RFS |
| Uzunlar et al, 2005 | 57 (77.1) | >1 | 35 | ✓ | ✗ | ↓ DSS |
| Zigeuner et al, 2004 | 184 (70.7) | >1 | 22.8 | ✓ | ✓ | ↓ RFS |
| Kim et al, 2004 | 318 (100) | >15 | NR | ✓ | ✓ | ↓ DSS |
| Uchida et al, 2002 | 112 (78) | >1 | 13.4 | ✓ | ✓ | ↓ OS |
| Olumi et al, 2001 | 48 (100) | >10 | 51 | ✗ | ✗ | — |
| Ljungberg et al, 2001 | 99 (74) | >5 | 19 | ✓ | ✗ | ↓ c DFS |
| Girgin et al, 2001 | 50 (62) | >20 | 20 | ✓ | ✗ | ↓ DFS |
| Haitel et al, 2001 | 104 (100) | >5 | NR | ✓ | ✗ | ↓ DFS |
| Haitel et al, 2000 | 97 (100) | >5 | 36 | ✓ | ✗ | ↓ DFS |
| Rioux-Leclercq et al, 2000 | 66 (NR) | >20 | 17 | ✓ | ✗ | ↓ DFS |
| Sejima et al, 1999 | 53 (NR) | NR | 2 | ✗ | ✗ | — |
| Vasavada et al, 1998 | 39 (71) | >1 | 0 | ✗ | ✗ | — |
| Sinik et al, 1997 | 39 (100) | >10 | 17.9 | ✓ | ✗ | ↓ OS |
| Papadopoulos et al, 1997 | 90 (NR) | Any positive | 33 | ✗ | ✗ | — |
| Gelb et al, 1997 | 52 (100) | >5 | 2 | ✗ | ✗ | — |
| Shiina et al, 1997 | 72 (NR) | >10 | 40.3 | ✓ | ✓ | ↓ OS |
| Moch et al, 1997 | 50 (100) | NR | 16 | ✓ | ✓ | ↓ OS |
| Hofmockel et al, 1996 | 31 (NR) | >1 | 16 | ✗ | ✗ | — |
| Lipponen et al, 1994 | 123 (NR) | Any positive | 33 | ✓ | ✗ | ↑ RFS |
| Kamel et al, 1994 | 56 (NR) | >1 | 11 | ✗ | ✗ | — |
| Bot et al, 1994 | 100 (74) | >50 | 32 | ✗ | ✗ | — |
| Uhlman et al, 1994 | 175 (NR) | >1 | 28 | ✓ | ✓ | ↓ DFS |
a Analyzed as continuous variable.
b In patients with metastases/advanced disease.
Interestingly, in multiple studies, p53 overexpression was not associated with TNM stage or grade, suggesting that as a marker it may provide information that is independent from conventionally acquired pathologic information.
Many studies have assessed the prognostic value of p53 on oncologic outcomes. Most recent and often better designed studies found an independent prognostic value of p53 regarding different survival outcomes (recurrence-free survival, disease-specific survival, and overall survival) (see Table 1 ). However, some earlier and usually smaller studies did not find a correlation between p53 immunoreactivity and outcomes, which might be due to lack of statistical power to detect a difference, due to cutoff or assay used, as well as a possible true nonsignificant effect. Further explanations might be the inclusion of nonclear histology, leading to high heterogeneity and nonspecific results. For correct data interpretation, it is also important to consider publication bias in this scenario because studies without significant results are probably not published as frequently as their positive counterparts. An overview of the published results regarding p53 expression and its impact on oncologic outcomes is provided in Table 1 .
Mutational analyses of p53 are another important avenue for evaluating p53 as a marker for ccRCC. A limitation of Immunohistochemistry is that it provides no information about wild-type or mutated protein status, and although immunohistochemical expression is often used as a surrogate for p53 mutational status, this is not entirely similar information. The frequency of p53 mutations is described between 0% and 44% overall. In most studies, single-strand conformation polymorphisms were evaluated and contained mostly the core domain between exon 4 and 8 or 5 and 8 because this is the most common site of p53 mutation. Still, up to 15% of p53 mutations occur outside of the core domain and suggest underestimation of mutational status in some studies. With the large differences in mutational status and protein expression, mutational analysis has not proven its utility for prognostication of outcomes yet.
It should also be noted as in later discussion that many analyses use more than one marker in combination. Many of these analyses include p53 as a key marker associated with cell cycle.
p21
p21 or cyclin-dependent kinase inhibitor 1 prevents cyclin-dependent kinases from phosphorylation of protein substrates and acts downstream of p53 regulation where it is activated by wild-type but not mutant p53. Therefore, p21 mainly acts as a tumor suppressor by blockage of cell proliferation as well as promotion of apoptosis, and loss of p21 can lead to uncontrolled cell growth. However, it seems that p21 can act as an inhibitor of the cell cycle as well as a growth permissive and harbors proapoptotic and anti-apoptotic properties, which are p53 dependent as well as p53 independent. Mutations in the gene locus of p21 are thought to be rare in renal cell carcinoma (RCC).
The cutoff for normal p21 expression assessed via immunohistochemistry was suggested to be greater than 30%. In one study assessing different histologic subtypes, it was also shown that positive nuclear and cytosolic staining for p21 was lower (median expression of 20% in nuclear assays) to minimal (median 0% in cytosolic assays) in ccRCC in comparison to other tumor subtypes. Interestingly, when comparing primary and metastatic tumor tissue, nuclear expression of p21 was higher and cytosolic expression was lower in the metastatic tissue. p21 expression was not associated with grade or stage when evaluated for this endpoint. Furthermore, high levels of p21 indicated poor prognosis in patients with metastatic disease, indicating mechanisms of therapy resistance associated with this status. p21 protein expression with regard to prognosis was evaluated in multiple studies, and according to its function in the cell cycle, patients with high and therefore intact levels of p21 demonstrated favorable disease-specific survival in some larger studies and was an independent predictor of this endpoint in patients with organ-confined disease as well. However, in previous and usually smaller studies, this association could not always be demonstrated. An overview of the published results regarding p21 expression and its impact on oncologic outcomes is provided in Table 2 .
| Study | No. of RCC (% ccRCC) | Cutoff for Alteration, % | Alteration Rate, % | Prognostic Value in UVA | Prognostic Value in MVA | Endpoint with Prognostic Value |
|---|---|---|---|---|---|---|
| Weber et al, 2014 | 145 (100) | <32.5 | 63 | ✓ | ✓ | ↓ DSS |
| Weber et al, 2013 | 132 (100) | <32.5 | 36.6 | ✓ | ✓ | ↓ DSS |
| Dahinden et al, 2010 | 527 (100) | NR | NR | ✗ | ✗ | — |
| Klatte et al, 2009 | 170 (100) | NR | NR | ✗ | ✗ | — |
| Weiss et al, 2007 | 366 (93.4) | <32.5 a | NR | ✓ | ✓ | ↓ DSS |
| Haitel et al, 2001 | 104 (100) | <10 | 42.3 | ✗ | ✗ | — |
| Aaltomaa et al, 1999 | 118 (NR) | NR | NR | ✗ | ✗ | — |
p27
Another cyclin-dependent kinase inhibitor is p27. p27 inhibits cyclin-dependent kinase 2 and leads to cell cycle arrest in the G1 phase. Therefore, similar to p21 to which it is structurally related, loss of p27 leads to uncontrolled cell cycle progression and cell division as well as tumor growth. However, transcription of the p27 gene is not controlled by p53 as it is in p21.
Depending on the cutoff used, p27 is detectable in around 60% of ccRCCs via immunohistochemistry. Loss of p27 was seen in tumors with higher Fuhrman grade, larger tumor size, and higher TNM status. Conversely, nuclear expression levels of p27 were lowest in benign tissue, higher in primary ccRCC tissue, and highest in metastatic ccRCC tissue, which seems contradictory to the known mechanism of action of p27. However, other study groups as well found higher levels of nuclear p27 in tumor tissue than in, for example, oncocytomas, chromophobe, or benign tissue, whereas papillary RCC demonstrated slightly higher expression rates. Most studies, however, found low nuclear expression of p27 to be associated with unfavorable oncologic outcomes. Still, in some studies, no correlation of p27 expression with oncologic outcomes was found. No found correlation may be partly attributed to heterogeneity in p27 expression in tumor tissue because one study found that low p27 expression at the border of invasion held prognostic significance for survival endpoints, while it did not when measured within the primary tumor.
These studies, however, focused on the nuclear staining of p27. One group recently focused on the importance of cytoplasmic expression of p27. The investigators found that high cytoplasmic staining for p27 was associated with unfavorable cancer-specific survival. However, the importance of this finding is not clear yet because the investigators defined high cytoplasmic expression as a higher expression in tumor tissue than matched benign tissue from other parts of the kidney and low expression as similar or lower expression of cytoplasmic p27 in tumor tissue than in benign tissue. It is unclear how the relation of p27 expression in benign and tumor tissue has to be in order to have a significant impact on prognosis.
An overview of the published results regarding p27 expression and its impact on oncologic outcomes is provided in Table 3 .
| Study | No. of RCC (% ccRCC) | Cutoff for Alteration, % | Alteration Rate, % | Prognostic Value in UVA | Prognostic Value in MVA | Endpoint with Prognostic Value |
|---|---|---|---|---|---|---|
| Kruck et al, 2012 | 140 (100) | Greater than corresponding benign tissue b | 24.3 | ✓ | ✓ | ↓ DSS b |
| Sgambato et al, 2010 | 125 (80) | <20 | 45.5 | ✓ | ✓ | ↓ RFS, ↓ OS |
| Dahinden et al, 2010 | 527 (100) | NR | NR | ✓ | ✗ | ↓ OS |
| Klatte et al, 2009 | 170 (100) | NR | NR | ✗ | ✗ | — |
| Liu et al, 2008 | 482 (87.9) | <40 | 82.2 | ✓ | ✓ | ↓ RFS, ↓ DSS |
| Pertia et al, 2009 | 52 (100) | No staining | 30.8 | ✓ | ✓ | ↓ RFS, ↓ DSS |
| Merseburger et al, 2007 | 251 (NR) | <5 | 46.2 a | ✓ | ✓ | ↓ DSS a |
| Pertia et al, 2007 | 52 (100) | No staining | 30.8 | ✓ | ✓ | ↓ RFS, ↓ DSS |
| Langner et al, 2004 | 171 (75.4) | <50 | 64 | ✓ | ✓ | ↓ RFS |
| Hedberg et al, 2003 | 218 (80.3) | <5 cells/core | 24.8 | ✓ | ✓ | ↓ DSS |
| Anastasiadis et al, 2003 | 154 (NR) | NR | NR | ✗ | ✗ | — |
| Hedberg et al, 2002 | 79 (83.5) | <60 | 29 | ✓ | ✗ | ↓ DSS |
| Migita et al, 2002 | 67 (100) | <50 | 31.3 | ✓ | ✓ | ↓ DSS |
| Haitel et al, 2001 | 104 (100) | <70 | 75 | ✓ | ✓ | ↓ RFS |
a In the invasion front tissue.
Related posts:
Biomarkers in Urologic Cancer
Urinary Biomarkers for Prostate Cancer
Current Status of Urinary Biomarkers for Detection and Surveillance of Bladder Cancer
Emerging Bladder Cancer Biomarkers and Targets of Therapy
MicroRNAs in Testicular Cancer Diagnosis and Prognosis
The Emerging Role and Promise of Biomarkers in Penile Cancer
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
