Expert Explains Effect of Age and Gender on Outcomes in RCC

Rana R. McKay, MD, sheds light on the impact of age and gender on the outcomes of patients with RCC, the implications of clinical determinants on toxicity, and where future research efforts will focus.

Understanding the implications of age and gender upon safety and efficacy outcomes in patients with renal cell carcinoma (RCC) who are receiving targeted treatment options is critical to the delivery of individualized care, according to Rana R. McKay, MD.

“It is important to identify the clinical determinants that impact outcomes in patients with RCC. Physiologic, genetic, and environmental differences between individuals of different ages and genders can play a role in the toxicity associated with treatment,” said McKay. “Developing strategies to optimize the modifiable parameters and enhance patient selection for therapy will be important for improving outcomes.”

To this end, McKay and colleagues conducted a pooled analysis of patients with metastatic RCC who had been treated on phase 2 and 3 clinical trials to investigate the impact of age and gender on survival. Patients were stratified by age: young (<50 years), intermediate (50-70 years), elderly (>70 years), and gender.

In total, investigators identified 4736 patients with metastatic RCC to include in the analysis. overall, no difference in overall survival (OS) was observed when stratified by age or gender. Specifically, OS outcomes were 21.0 versus 17.3 months for elderly versus intermediate age groups (P = .382), respectively, and 20.0 versus 17.3 months for young versus intermediate age groups (P = .155), respectively. The OS outcomes for gender were 19.8 versus 19.0 for male versus female patients (P = .510), respectively. Interestingly, progression-free survival (PFS) was shorter in younger individuals compared with the intermediate age patients (6.0 vs 7.1 months, P <.001), but similar across gender groups. Although all-grade adverse events (AEs) were more common in elderly patients, serious AEs were similar between groups.

In an interview with OncLive, McKay, assistant professor of medicine, medical oncologist, and head of a multidisciplinary prostate cancer clinic, at the University of California, San Diego, shed light on the impact of age and gender on the outcomes of patients with RCC, the implications of clinical determinants on toxicity, and where future research efforts will focus.

OncLive: What is some of the evidence available to suggest that age and gender play a role in cancer outcomes?

McKay: There is a growing body of evidence suggesting that age and gender play a role in cancer outcomes; multiple factors exist for this. The factors could be biologic, meaning different genetic profiles in men and women, as well as in younger and elderly individuals. There are also different pharmacokinetics; for example, how we process different drugs. Drugs may be processed differently in men versus women [or based on] different ethnicities and the presence of different enzymes that help with drug metabolism.

We know that toxicity patterns are different between younger and older individuals, as well as between male and female patients. That was the impetus behind doing this study: to really tease out, in the context of a large, pooled analysis of clinical trial data, outcomes based on age and gender.


What inspired you to look at this more closely in RCC with targeted agents?

The reason to look at this in RCC is that there has never really been a deep dive into these gender and age differences in patients receiving targeted therapies. We wanted to home in on understanding whether patients have differential outcomes based on age, whether patients have differential outcomes based on gender, and tease that out in the context of a very large dataset. Prior to this effort, this has not really been looked at. Every clinical trial includes age and gender as baseline characteristics, but [we really wanted to look] at this in aggregate to determine whether nuances exist with regard to how patients may or may not respond or [whether their] outcomes [differ] between these categories.

What were the methods used to perform this analysis?

We conducted a pooled analysis of patients with metastatic RCC who were treated on phase 2 or phase 3 clinical trials examining targeted therapy agents; these patients received VEGF TKIs, monoclonal antibodies to VEGF, mTOR inhibitors, and interferon. We stratified patients based on age, and we used the cutoffs of <50 years, 50-70 years, and >70 years to define our age category of young, intermediate, and elderly, respectively. Then, we stratified by gender, specifically male and female. We conducted Cox regression analyses to adjust for multiple factors that could inform outcomes from sites of metastases, like bone and liver metastases, International Metastatic RCC Database (IMDC) risk criteria, and other prognostic variables, to tease out whether age and gender impacted outcomes in this large dataset.

How did you select which trials to include in this analysis and what did the patient populations look like in the ones selected?

The trials that were included were those examining targeted therapy agents, that were conducted in patients with metastatic or advanced RCC. All these studies were [done in patients with] locally advanced or metastatic RCC and all of them were examining TKI therapy or other targeted therapy agents; that was a homogenizer. We did include trials of both first- and second-line agents and we, therefore, conducted subset analyses based on the line of therapy. We also conducted subset analyses based on IMDC risk stratification, so favorable, intermediate, and poor. Additionally, we conducted subset analyses based on the type of therapy, so whether it was VEGF, mTOR, or interferon based. Of course, we looked at the overall cohort, as well.

What were the findings?

In total, this was a very large database. We identified 4736 patients with metastatic RCC. For the overall cohort analysis, when we look at the total population no difference in OS was observed when stratified by age nor gender. The OS, when stratified by age, was 21 months for the elderly versus 17.3 months for the intermediate-age group. In the younger group, OS was 20 months versus 17.3 months and the P values were not found to be statistically significant. Similarly, no difference was observed [when stratified] by gender [either], when we look at the overall study cohort without doing the subset analyses. For males, compared with females, the OS was 19.8 months versus 19.0 months, respectively.

What we did see was a signal that PFS was shorter in younger individuals compared with the intermediate-age group. Specifically, PFS was 6 months [in the younger age group] versus 7.1 months in the intermediate-risk group; however, PFS was similar across gendered categories. That poses the questions, “Do younger patients inherently present with a more aggressive disease phenotype? Are they more at risk for hereditary RCC cancer syndromes?” The biology behind that still needs to be teased out but there was a signal indicating that potentially younger patients may fare worse and that could be related to disease biology.

It was interesting to see that the outcomes did not differ for the elderly individuals, because in the context of patients enrolling in clinical trials, these trials require patients to have their comorbidities to be controlled. Our elderly patients actually did not have inferior outcomes. However, what we did see was all grade AEs were more common in the elderly, and that included fatigue, diarrhea, decreased appetite, and weight loss. Those events appear to be a bit more common in our elderly population in this study. However, when we look at serious AEs, grade 3/4, they were similar across the age and gender categories. As such, it’s just those grade 1/2 events that seem to be more common in the elderly patients.

Results showed that PFS was shorter in the younger patient population compared with intermediate-age patients. Why do you think that is?

With regard to the results of the study, I think they were surprising. When we were looking at the data, our original hypothesis was that the elderly patients would do worse than the younger patients. We thought the younger patients would have more robust outcomes. To go back to the methodology, we used the intermediate age category as our reference point. We compared elderly to intermediate age and young to intermediate age, so we could understand the differences between those extremes of age. Initially, we thought the elderly patients would have inferior outcomes, but it seems that they did not in our dataset. Maybe there’s a signal that the younger patients may have a shorter PFS. However, ultimately, as a whole, the study showed that no difference, but we need to understand that underlying disease biology of what, on a molecular level, may be driving [these] differences between patients.

What are the implications of these findings?

With regard to the implications, [this analysis] warrants a very open discussion between a patient and their clinician regarding risk assessment for someone being on a clinical trial; this actually speaks more to toxicity management. We did see more toxicity in our elderly patients, so we need to ensure that systems are in place to closely follow individuals and mitigate toxicity early so [that those events] do not become severe; this is very important. The other point is that we know that there are pharmacokinetic and pharmacodynamic differences between agents. Patients may need dose modifications to be maintained on a stable dose and being keen on watching for toxicity so we can modify doses; that is really the big take-home message from this research.

Understanding outcomes by age and gender is critical to the delivery of individualized and multidimensional patient care. The purpose of this study was to look at outcomes on the basis of age and gender to understand how these determinants can impact outcomes. Largely, they do not necessarily impact cancer-specific outcomes but they can potentially impact toxicity and having that ongoing discussion between the patient and provider to ensure that patients are aware of how they should be taking their medications, that they are reporting toxicity, and clear instruction regarding dose modification is in place if that is needed to handle events; that is the biggest thing.

How do you approach these discussions in practice?

It is about having more open dialogue and discussion with patients about outcomes. It is also saying that not necessary to discriminate the choice of therapy based on age or gender. Elderly patients did not do worse, and so having a discussion where you share with them what all their treatment options are, all the pros and cons of each of those options, the toxicity profile of each option, and coming to an agreement together on what the best path forward is.

Are there any unanswered questions left by this analysis that should be explored with future research?

One of the biggest questions has to do with immunotherapy. In this dataset, we did not have a large cohort of patients who were receiving immune checkpoint blockade; that is going to be the next layer to add on here. We started some preliminary data that there may be differential outcomes based on gender for immunotherapy; whether or not that is true is yet to be determined. Would there be a biologic rationale for that? Is the immune milieu different in younger versus elderly patients? Could that potentially impact response to immune checkpoint blockade? That is the next layer [to this research].

Was there anything else you’d like to highlight?

Shifting gears, the COVID-19 pandemic has impacted just about every single sector of society and largely impacted the way that we deliver care for our patients with cancer. We are in this data-free zone right now where many decisions around care delivery for our patients have been based on indirect scientific data, anecdotal data, or expert opinion. We are [facing] a knowledge gap between how best to augment therapy for our patients and understanding the risk for our patients. Some of the initial reports about COVID-19 in cancer are that patients with cancer seem to be particularly susceptible to the complications associated with the virus, both mortality and morbidity.

We have developed a large, grassroots registry called CCC19 that is now comprised of greater than 100 largely academic, as well as community oncology groups, seeking to answer some of these critical questions with regard to understanding how patient and disease determinants impact COVID-19–specific outcomes and vice versa. How does COVID-19 impact cancer outcomes, toxicity with therapy, and response to therapy? This is a huge effort with large institutions.

Jeremy L. Warner, MD, MS, of Vanderbilt University Medical Center, initiated the registry but many institutions have gotten on board. We are hoping that this registry will be able to be very nimble and answer these questions in a very quick way and disseminate the information to the community so that we can understand how best to mitigate practice for our oncology patients.

Reference

  1. Panian J, Lin X, Simantov R, et al. The impact of age and gender on outcomes of patients with advanced renal cell carcinoma treated with targeted therapy. [published online ahead of print, Mar 16, 2020]. Clin Genitourin Cancer. doi:10.1016/j.clgc.2020.03.010