ASCO GU 2020: Current Triggers for Intervention in Patients on Active Surveillance

As clinicians, it is important to understand the spectrum of the disease biology, namely we need to act on the masses that may get away from us and surveil the masses that are not going to harm the patient. The harms of overtreatment are well documented and include psychological stress, financial toxicity, loss of kidney function, an unknown survival benefit, surgical complications, and unnecessary care. As such, we must balance risks of treatment with the risks of active surveillance, accounting for the competing risks of death. Work by Dr. Kutikov’s group has looked at quantifying competing risks of death, taking into account prognostic variables such as race, age, tumor size, age, and Charlson comorbidity index.¹ But to Dr. Kutikov’s own admission, using post-treatment data to make no-treatment decisions is imperfect and clinical gestalt still plays a large role. Other risk calculators are available, most notably the NSQIP surgical risk calculator available here.

Ultimately, the risk of active surveillance is the risk of metastatic progression. The first factor involved in this risk is tumor histology. We know that mass anatomic complexity can be harnessed to risk stratify for malignancy (model AUC 0.78) and high-grade disease (model AUC 0.73). Patients selected for active surveillance have lower RENAL nephrometry scores with tumors that are smaller, further from the sinus or urothelium, more often polar, and less often hilar. Renal mass biopsy has excellent utility for differentiating benign from malignant tumors, however, biopsy has poor utility for differentiating low from high-grade tumors. Furthermore, renal mass biopsy is underutilized with only 2-64% of patients having a renal mass biopsy that are going on active surveillance. Ultimately, 30% of partial nephrectomies are still being performed for benign masses in the United States.²

The second factor associated with risk of metastatic progression is tumor size. For each one cm increase in size, there is a ~15% increased risk of malignancy and a ~25% increase in high-grade disease. Ultimately, metastases typically occur for patients with renal masses >3 cm and the safety of active surveillance is best in T1a tumors (and even better in <3 cm tumors). Highly selected cohorts of patients with cT1b/T2 masses may be observed with progression rates over ~3-year follow up ranging from 0-6%.

The third factor associated with the risk of metastatic progression is tumor growth. Linear tumor growth is the most often metric reported and constitutes the maximal tumor diameter and measured in changes in mm or cm per year. However, Dr. Kutikov notes that the estimated tumor volume change may be a better surrogate for tumor biology as he reminds us that a two cm tumor is made up of ~8 billion cells and a three cm tumor is made up of ~28 billion cells. The most commonly used formulas are 0.523*x*y*z or 0.523*x*y*(x+y/2) or 0.523*x.³ The linear growth rate in pooled analyses is ~2.2 to 3 mm/year, which is ~1 mm/year in prospective registries such as RCCC Canada and DSSRM. Additionally, 10-30% of masses have zero order or negative growth kinetics with essentially no metastatic progression. A linear growth rate of ~8 mm/year occurs in the <3% of patients who progress to metastatic disease for cT1a masses. Dr. Kutikov cautions that we should be aware that rapidly growing masses can metastasize, but may still be an oncocytoma. Looking further at the Fox Chase Cancer Center cohort of active surveillance patients, over a follow-up of at least five years (median 67 months), there were 457 patients with 544 lesions and a median size of 2.1 cm.³ Interestingly, patients associated with delayed intervention were more likely to have tumor growth, of which 1.7% of patients metastasized. Dr. Kutikov reiterates that linear growth rate >5mm/year forces the clinicians hand.

Ultimately, 4-65% of patients will progress to definitive therapy, with a very low rate of metastatic progression of 0.7-7% (3% in pooled analysis). Additional factors that must be considered are the impacts on quality of life. Several studies have shown that some quality of life metrics worsen over time, including illness uncertainty associated with worse quality of life outcomes. The appropriate management of illness uncertainty is likely an important aspect of active surveillance. In fact, the DSSRM registry has a structured active surveillance program showing that mental health was not adversely affected while on surveillance and actually improves over time. The following algorithm is what Dr. Kutikov uses for his active surveillance program:



Dr. Kutikov concluded with the following remarks regarding active surveillance for small renal masses:

• Contemporary active surveillance remains a process of shared decision-making
• Triggers for intervention include positive growth rate (especially >5 mm/year) and tumor size >3-4 cm
• Patients with <3 cm tumors and no/slow growth are likely the best candidates for active surveillance