 I'd like to thank the organizing committee for allowing me the opportunity to speak this morning. So I thought this was best, the issues of comorbidity and renal dysfunction in the management of localized kidney cancer are best illustrated with a case example. So this is a 77 year old female who presents with a 3.2 centimeter left renal mass which was incidentally found. In clinic she appears quite frail and is diabetic, hypertensive and has arthritis but her serum creatinine is normal at 0.89. So there are two possible outcomes for a patient with a kidney tumor. Either they can die of kidney cancer or they can die of other causes. Factors that lead to death from kidney cancer include tumor characteristics such as tumor size, stage and grade. And factors that influence other causes of death are the patient's age and comorbidity. There's some evidence that chronic kidney disease may lead to death from other causes, mostly through cardiovascular morbidity and I'll discuss that later. And if you treat the tumor, presumably you would decrease the risk of dying of kidney cancer. Sorry, I don't know what happened to the slides there. Okay, I apologize again. Okay, so here we go. This is where we were. So if we treat the kidney cancer we could potentially decrease the risk of dying of kidney cancer. We may at least in the short term increase the risk of dying of other causes mostly through postoperative mortality and you may indirectly affect death from other causes through its effect on chronic kidney disease. So first let's talk about death from kidney cancer. So we know that the malignant potential of a tumor is directly related to its size such that the larger the tumor, the higher the five-year cancer specific mortality if it is treated. We can enter her information into this risk calculator such as this one published by Lane at Allet at the Cleveland Clinic and you see for a female with no local symptoms and no history of smoking, for a tumor of 3.2 centimeters in size at her age her risk of having a malignant tumor is estimated to be 78.1%. We can use this Nomegram which is renal nephrometry based published by Kudakov in European Neurology and her nephrometry score is 8a and so this gives us an estimate of what is likelihood of having a high grade disease so likely to kill her and this is estimated to be 24% using this Nomegram. What about her risk of dying of other causes? Well if we use the Social Security Actuarial Life Table for a 77-year-old female her life expectancy is about 11.5 years and we also know that the likelihood of dying of other causes is directly related to her comorbidity status so for each increase in Charleston comorbidity score you have a corresponding increase of risk of death from other causes whereas kidney cancer death is not related to comorbidity. This graph illustrates the chances of dying of kidney cancer in red versus chance of dying of other causes in blue as it relates to tumor size as well as comorbidity. So you see that for a tumor less than 4 centimeters in size and a Charleston score of one she's much more likely to die of other causes than of kidney cancer. So if we use the competing risks model again from Kudukov and Uzo's group at Fox Chase you see that at five years her risk of dying of other causes is estimated to be 13.1% and her chance of dying of kidney cancer if she is treated is estimated to be 3.2% of five years. So with that we present her treatment options radical nephrectomy, partial nephrectomy, tumor ablation active surveillance with or without renal mass biopsy. So what about the influence of renal function and how does that affect our decision making? Well we can estimate her GFR using the MDRD formula which gives us an estimated GFR of 64.5 and we know that partial nephrectomy will preserve more kidney function than radical nephrectomy both in the short term as well as in the long term. We know that chronic kidney disease is related to all-cause mortality with an increase in all-cause mortality with subsequent decreases in GFR and retrospective studies comparing partial to radical nephrectomy using serum Medicare have demonstrated a survival advantage and it was hypothesized that this was due to better renal functional preservation and the resulting decreased cardiovascular morbidity. And if you look at the effects of the postoperative GFR there's a direct relationship with overall survival such that for a lower postoperative GFR you have a lower overall survival as well as a lower cancer cardiac specific survival. That being said however there's an interesting study which was published this year in cancer and what they did is they looked at partial nephrectomy radical nephrectomy and non-cancer controls as well as non-muscle invasive bladder cancer and they compared the survivals between these treatments. And what they showed is for radical nephrectomy there was no difference in survival compared to non-cancer controls suggesting that radical nephrectomy does not affect overall survival and they compared partial nephrectomy to non-cancer controls and they showed that partial nephrectomy patients actually did better than non-cancer controls. And since there's no biological explanation for this they hypothesized that this was due to selection bias and unmeasured confounders. And when partial and radical nephrectomy were evaluated prospectively in the European trial as we know partial nephrectomy did not do better than radical if anything it did a little bit worse although this study was criticized for being underpowered. And in the soon to be published renal functional results from that study you see that although radical nephrectomy patients did have a slightly higher post-operative creatinine on average it did not deteriorate over time and if anything their serum creatinine appeared to improve over time. And although there were differences at GFRs of 60 and 45 when for lower GFRs the more severe chronic kidney disease there was very little difference between radical nephrectomy and nephron sparing surgery. Also not all chronic kidney disease may be created equal. Lane et al recently published this in the Journal of Urology and they showed that surgically induced chronic kidney disease may not be as bad as medically induced chronic kidney disease. They showed that patients with surgical CKD had similar survival to patients with no CKD whereas those with medical CKD had worse overall survival as well as a worse deterioration in renal function post-operatively. Also we can't ignore the impact of morbidity of surgeries particularly in the elderly. And although there is an increase in complication rate with age for both radical and partial nephrectomy there is a more dramatic increase in complications with partial nephrectomy such that as an octogenarian has a 2.4 time risk of having a complication after partial nephrectomy compared to a patient under the age of 50. Also the 30 day mortality is not insignificant. So for example for a patient over the age of 80 their estimated 30 day mortality is estimated to be 3.6%. Also we can enter her preoperative characteristics into this American College of Surgeons Nesquip Surgical Risk Calculator to get a sense of her postoperative complication risk. So for our patient with a laparoscopic let's say radical nephrectomy her risk of having a complication is about 10%. Risk of having a serious complication is about 9%. Risk of death is small at less than 1% and she's predicted to be in the hospital for about two and a half days. So we discussed the pros and cons of all of the approaches with this patient and she decided to be placed on initial active surveillance. At six months we repeated the CT scan and this showed an increase in the size of the tumor from 3.2 centimeters to 3.9 centimeters. What does this increase in size tell us about the malignant potential of the tumor? Well as Dr. Jewett presented this morning in a prospectively followed cohort of patients who were untreated or placed on active surveillance for those who had renal mass biopsies although malignant tumors started off at larger sizes the growth rates on average paralleled those of benign tumors. That being said there was some concern for an aggressive high grade disease and therefore the patient underwent an uncomplicated laparoscopic radical nephrectomy. Pathology revealed a 3.6 centimeter renal cell carcinoma clear cell type Furman grade two. She was discharged on postoperative day number three to a skilled nursing facility. At six months she's doing well with a creatinine of 1.2 and an estimated GFR of 46. So how can we put all of this together? Well ideally we take all of the best published data and put it in a large model like a Markov state transition model. We then take our patient through all possible treatment options. We can then model certain transitional probabilities such as probability of cancer recurrence of dying of kidney cancer, chronic renal sufficiency and so on. And our model can give us what it estimates to be the best treatment in terms of life expectancy as well as quality adjusted life expectancy. So in summary, patient age, comorbidity and medical CKD appear to influence the risk of death from other causes in patients with small renal masses. What is needed however is a more accurate estimate of death of kidney cancer in untreated patients since most of the data that presents survival in patients with kidney cancer are in treated patients. In addition both cancer and non-cancer risks should be taken into account when considering treatment options in patients with localized kidney cancer. Thank you.