Elective partial nephrectomy (PN) is now commonly advised for patients with renal masses suspicious for renal cell carcinoma whenever feasible, and this appears to be very sensible for small renal tumors.
But should this also apply to larger tumors, such as clinical T1b/T2a (4-10 cm) in the presence of a normal contralateral kidney? Stated another way, has the pendulum swung too far in favor of PN? This has now become one of the controversial topics in our field, based on a variety of interesting studies over the past few years.
Why avoid radical nephrectomy?
To sort this out, however, we should start with a more fundamental question: Why have we felt so strongly about avoiding radical nephrectomy (RN)? There is a dogma in our field that has proved very persuasive, as follows: RN is associated with an increased incidence of chronic kidney disease (CKD), and this inevitably leads to an increased risk of morbid cardiovascular events and accelerated mortality rates. How strong is the evidence that supports this paradigm, or is it in reality a house of cards?
Most of the evidence supporting this dogma is derived from retrospective studies, which are likely contaminated by selection bias. The majority of these studies utilized sophisticated statistical techniques to adjust for recognized and even unrecognized biases, but there is no way to eradicate this concern.
A recent meta-analysis of this literature provides the most damning evidence. This study looked at more than 30 retrospective studies of PN vs. RN, and the results showed that PN offered an advantage in terms of better renal function and overall survival (19 percent advantage for PN). PN was also associated with more perioperative morbidity, although most of this was readily manageable.
Selection bias may influence outcomes
All of this makes sense. However, the authors also reported that PN had an advantage in terms of better cancer-specific survival, with a 29 percent difference! How can this be? PN is a great procedure, but it is clearly not a stronger oncologic intervention than RN. The only reasonable way to explain this result is selection bias, and we cannot ignore the strong possibility that selection bias is also driving the other results in favor of PN. How do we prevent selection bias?
The best way to eliminate selection bias, of course, is a randomized clinical trial, and such a trial was published in 2011, namely EORTC 30904. This provocative trial randomized a large number of patients (n = 541) with small renal tumors (< 5.0 cm) and a normal contralateral kidney to either PN or RN.
As expected, the trial showed that PN offered an advantage to RN in terms of better renal function, and there was slightly increased perioperative morbidity in this arm. Based on the current paradigm, we anticipated better overall survival in the PN group, but this was not found. In fact, the 10-year overall survival was actually better in the RN group (81 percent vs. 76 percent, p = 0.03), and cardiovascular events were more common in the PN group, although this did not reach statistical significance.
An uncomfortable finding
Needless to say, these results were highly unexpected, and many would like to sweep this study under the carpet – it is very uncomfortable when our common belief systems are challenged. This trial has some flaws, but it would be intellectually dishonest to ignore it. No matter how one would like to interpret it, one thing is certain: The renal functional advantage of PN in the setting of a normal contralateral kidney may not be as beneficial as previously believed.
We recently pursued this and developed a hypothesis that there may be a difference between CKD due to surgical removal of nephrons (CKD-S) compared with that due to medical comorbidities (CKD-M). Patients with CKD due to diabetes or other medical diseases will continue to suffer from these comorbidities, and the medical literature is very convincing in showing that such patients will continue to experience declining renal function (about 3 percent to 5 percent per year), eventually leading to cardiovascular events and other adverse sequelae. We postulated that patients with CKD-S could be more stable, because they typically do not need more surgery. Stated another way, the driver toward CKD is no longer present in these patients.
To study this hypothesis, we looked at more than 4,000 patients with renal tumors who were managed with either PN or RN at our center during the past decade. This cohort included more than 1,000 patients with CKD-M (pre-existing CKD, although they should be designated CKD-M/S because they then had surgery, too) and almost 1,000 patients who developed CKD-S.
We found that patients with CKD-M/S experienced an average annual decline of renal function of 4.7 percent, consistent with other reports in the literature, while patients with CKD-S were much more stable (annual decline in renal function of only 0.7 percent). In addition, the survival of patients with CKD-M/S was substantially worse than for patients with CKD-S, who exhibited survival that was very similar to patients with no CKD at all.
We recently reported an important study that builds on this line of investigation by providing a control group of 42,000 patients with pure CKD-M (CKD due to medical causes and not needing surgery) and controlling for potential confounding factors such as comorbidities and new baseline GFR. This more sophisticated analysis again demonstrates that patients with CKD-S are much less likely to progress towards renal failure. In addition, non-renal cancer-related survival in the CKD-S group was substantially improved when compared with the CKD-M/S and CKD-M groups.
Less advantage for PN?
While further studies are required, these data suggest that the renal functional advantage of PN in the setting of a normal contralateral kidney may not be as beneficial as previously believed. The experience with donor nephrectomy, albeit an even more highly select patient population, also supports this assertion.
How do we reconcile all of this recent data? We still strongly believe in PN, particularly for patients who do not have a normal contralateral kidney, and for most patients with small renal tumors (T1a). In the former group, PN is essential and in the latter it should still be prioritized, because most of these tumors have limited oncologic potential, and RN is gross overkill.
The main controversy now centers on the patient with a clinical T1b/T2a renal mass and a normal contralateral kidney. These tumors have increased oncologic potential and have often already replaced a substantial portion of the functioning nephron mass, leaving less to be saved by PN. In these patients, it is not clear what the optimal form of management should be, and a randomized clinical trial will be needed to allow for more rational patient counseling in the future.
Dr. Campbell is Professor of Surgery and Vice Chair and Program Director of the Department of Urology in Cleveland Clinic’s Glickman Urological & Kidney Institute.
Dr. Fergany is a staff member of the Department of Urology in the Glickman Urological & Kidney Institute.