Now that the FDA has carved out nonmetastatic, castration-resistant prostate cancer as a new disease state, the decision about whether to administer recently approved antiandrogen therapies in this setting hinges on the rate of increase in prostate-specific antigen levels and comorbidities.
Leonard G. Gomella, MD
Now that the FDA has carved out nonmetastatic, castration-resistant prostate cancer (nmCRPC) as a new disease state, the decision about whether to administer recently approved antiandrogen therapies in this setting hinges on the rate of increase in prostate-specific antigen (PSA) levels and comorbidities, according to experts in the field.
The development of therapies for nmCRPC has been hailed as an advance for men with localized prostate cancer whose PSA levels continue to rise despite androgen deprivation therapy (ADT) and other treatment.
In February 2018, apalutamide (Erleada) became the first FDA-approved drug for nmCRPC, an indication that also marked the first time metastasis-free survival (MFS) was used as a primary endpoint in a clinical trial.1 Five months later, the agency broadened the indication for enzalutamide (Xtandi) to include patients with nmCRPC.2 (For both drugs, patients should also receive a gonadotropin-releasing hormone analogue concurrently or should have had bilateral orchiectomy). And in February 2019, a new drug application was submitted to the FDA for darolutamide, a novel androgen receptor agonist, in combination with ADT for patients with nmCRPC.3
The drug approvals open the door for new options for a sizable proportion of men with prostate cancer. The incidence of nmCRPC in the United States is estimated at 50,000 to 60,000 patients annually1; overall, nearly 165,000 men expected to be given a diagnosis of the disease in 2018.4
“These were men who had been treated for prostate cancer but had a slowly rising PSA and no evidence of any metastatic disease. Now having options to treat these patients with 2 agents that interfere with the androgen pathway is very helpful for us,” Leonard G. Gomella, MD, said in an interview with OncologyLive®.
“It used to be a watch-and-wait approach for many of these patients until something happened or showed up on a scan. It is great now that we have something that we can use early to reduce the PSA in many of these men and prevent the progression of their disease or at least delay it for a period of time,” said Gomella, who is director of the Kimmel Cancer Center Network, Thomas Jefferson University Hospital in Philadelphia, Pennsylvania.Broadly speaking, the FDA defines nmCRPC as a state of “rising prostate-specific antigen despite castrate levels of testosterone and no radiographic evidence of distant metastatic disease.”5 Patients with a short PSA doubling time (PSADT) of ≤10 months face a risk of developing metastatic disease and prostate cancer—specific death.6
The pivotal phase III clinical trials for apalutamide and enzalutamide, as well as a study into darolutamide, were conducted in similar patient populations. Key eligibility requirements included PSADT of ≤10 months from baseline despite maintaining castrate levels of testosterone (Tables 1 and 2).7-9 MFS was defined as the time from randomization to the time of first evidence of bone or soft tissue distant metastasis or death due to any cause, as confirmed by blinded independent central review.
In the SPARTAN study, investigators randomized 1207 men with nmCRPC to ADT with either apalutamide or placebo. The median MFS time for patients who received apalutamide (at 240 mg daily) was 40.5 months (95% CI, not estimable [NE]—NE) versus 16.2 months for those who took a placebo (95% CI, 14.6-18.4). That translated into a 72% reduction in the risk of metastasis or death with apalutamide (HR, 0.28; 95% CI, 0.23-0.35; P <.001).7
The PROSPER trial randomized 1401 men with nmCRPC and a PSA doubling time of ≤10 months to ADT with either enzalutamide and/or placebo. The median MFS for patients treated with enzalutamide (160 mg daily) was 36.6 months (95% CI, 33.1—not reached) versus 14.7 months (95% CI, 14.2-15.0) for those given a placebo. Those findings resulted in a 71% reduction in risk of metastasis or death with enzalutamide (HR, 0.29; 95% CI, 0.24- 0.35; P <.001).8
In the ARAMIS trial, 1509 patients were randomized 2:1 to ADT with either darolutamide or placebo. The median MFS was 40.4 months for patients who received darolutamide (600 mg twice daily) versus 18.4 months for those who took a placebo. Darolutamide therapy was associated with a 59% reduction in the risk of metastasis or death (HR, 0.41; 95% CI, 0.34- 0.50; P <.001).9Although the clinical trials produced MFS benefits for patients who received the therapies that target androgen receptor activity, the data assessing the secondary endpoint of overall survival (OS) with prompt treatment versus active surveillance have not yet matured. In SPARTAN, the median OS was not reached with apalutamide versus 39 months with placebo; there was a trend toward an OS benefit with apalutamide (P = .07).7 In PROSPER, the median OS was not reached with either enzalutamide or placebo; the hazard ratio for death with enzalutamide was 0.80 (P = .15).8 And in ARAMIS, darolutamide was associated with a lower risk of death than placebo (HR, 0.71; 95% CI, 0.50-0.99; P = .045).9 Other questions include whether patients whose cancers are less aggressive than those of the trial participants would be better candidates for active surveillance. Additionally, patients who choose treatment face immediate drug costs.
Despite these uncertainties, there are several objective factors that PROSPER and SPARTAN investigators use in recommending either immediate treatment or delay to the patients they see in clinical practice.
The most concrete among these considerations are patient comorbidities and the speed at which PSA levels are rising. Some comorbidities, such as a history of seizures or elevated seizure risk, may preclude the use of either enzalutamide or apalutamide. Both medications cross the blood—brain barrier, so patients with elevated seizure risk were excluded from both SPARTAN and PROSPER.
The apalutamide prescribing information carries a warning about the risk of falls and fractures, with the advice that patients should be evaluated for risk and treated with bone-targeting agents as appropriate.10 The enzalutamide label includes a warning about the risk of seizures, particularly in patients with predisposing factors.11
Such patients may soon have a different treatment option because they were eligible for inclusion in the ARAMIS trial evaluating darolutamide. Findings demonstrate that darolutamide was not associated with a higher incidence of seizures, falls, or cognitive disorder than placebo.9
Other comorbidities, such as hypertension and ischemic heart disease, do not preclude treatment with these therapies, but such conditions may increase the risk of serious adverse events (AEs).
“I discuss the options and pros and cons with the patient and individualize plans based on the patient general health, disease progression rate, and preferences,” Maha H. Hussain, MD, MBChB, FACP, FASCO, the lead investigator for the PROSPER trial, said in an interview. Hussain, a 2015 Giants of Cancer Care® award winner, is the Genevieve E. Teuton Professor of Medicine and the deputy director at the Robert H. Lurie Comprehensive Cancer Center of the Northwestern University Feinberg School of Medicine in Chicago, Illinois.
“For example, in situations where there is a rapidly rising PSA and no major comorbidities, I recommend consideration of therapy relatively sooner,” said Hussain. “On the other hand, for patients with significant comorbidities and particularly if they have a slowly rising PSA, I advise that their comorbidities are better controlled before starting therapy. Their disease isn’t progressing quickly, so there is no rush to treat them.”
Because of the use of PSA screening, less than 20% of men had imaging evidence of metastasis at the time of diagnosis, Luo et al reported in 2016.12 The standard of care for these nonmetastatic prostate cancers is surgery or radiation. These treatments can be curative, but patients are monitored for biochemical recurrence in the form of a rising PSA level, which often occurs before detectable metastases. If treatment is warranted because of PSADTs, the standard therapy has been ADT.12
Hussain reported that in her practice, about 9 of every 10 patients with nmCRPC have chosen early treatment since the new options became available last year. Other practitioners are seeing a more evenly distributed mix.
“The deciding factors in the patients I’ve treated seem to be how they feel about rising PSA levels and how they feel about the risk of adverse effects,” Julie N. Graff, MD, an investigator for the SPARTAN trial, said. “Some men suffer serious anxiety when they see their PSA levels rising. For them, it’s important to treat early because it greatly improves their mental health and thus their quality of life.
“Other men are already suffering pretty serious side effects from androgen deprivation therapy—and often other medications as well—and most of them delay treatment because they just don’t wish to risk more side effects that further reduce their quality of life,” said Graff, an associate professor of medicine at Oregon Health & Science University in Portland. “When you don’t have clear data about medical outcomes, quality of life is a perfectly reasonable basis for making decisions.”
Graff tries to avoid steering the vast majority of her patients in any particular direction. She does not think the data released to date justify any predictions on her part regarding what is likely to optimize outcomes for patients with nmCRPC. She therefore tries to give a neutral presentation of results to date and let patients decide. Other experts choose to interpret the available data and offer significant guidance.
“I absolutely recommend treatment for nmCRPC patients whose PSA doubling times are equal to or less than 10 months, correlating with the PROSPER, SPARTAN, and ARAMIS inclusion criteria for these trial populations,” said Neal D. Shore, MD, an investigator for both PROSPER and ARAMIS. “Assessing the MFS data, the adverse events reported, and patient quality-of-life information, initiating nmCRPC treatment is supported both by the clinical evidence and the known likelihood for biologic progression.
“The same conclusions have not yet been established for nmCRPC patients with higher PSA doubling times,” noted Shore, who is the medical director for the Carolina Urologic Research Center and Atlantic Urology Clinics in Myrtle Beach, South Carolina. “Data from other trials suggest that these patients will progress very slowly to metastatic disease when left untreated. Moreover, if these patients are reasonably monitored and their PSA doubling times shorten, there should still be an opportunity to initiate nmCRPC treatment prior to radiographic imaging positivity [for metastasis].
“It’s a patient—physician shared discussion in order to weigh the risk–benefit analysis of adverse events versus delay in MFS for the PSADT patients north of 10 months,” he added. “For that reason, I generally advise nmCRPC patients with slowly rising PSA levels to delay treatment.”
Any major shift to early treatment would constitute a significant change in prostate cancer care. That said, none of the physicians interviewed for this story thought such a shift would require much additional work for patients or providers. Patients who begin treatment on either enzalutamide or apalutamide typically visit 6 weeks later so that providers can look for signs of AEs and address them as needed. However, after 1 or 2 such checkups, the time between visits typically lengthens back to 3 months, which is also typical for patients with nmCRPC who opt for active surveillance.
“There may be a few extra visits associated with treatment, but neither drug requires the sort of appointment schedule that will tax patients or practices to the breaking point,” Graff said. “As for extra tests, there really aren’t any. We measure PSA levels every 3 months whether or not patients opt for treatment. It’s not like I’m going to need CT [computed tomography] scans every 3 months to see if the drug is working. PSA is a pretty good measure in this population.”When patients opt for treatment, physicians must then decide which drug to prescribe. Apalutamide and enzalutamide have not been tested head-to-head, so there is no definitive answer to the question of whether one is superior.
In terms of all-grade AEs, findings from SPARTAN showed that the rates of toxicities that were higher with apalutamide than with placebo, respectively, included rash (23.8% vs 5.5%), falls (15.6% vs 9.0%), fracture (11.7% vs 6.5%), and hypothyroidism (8.1% vs 2.0%). The rate of serious AEs was 24.8% for apalutamide compared with 23.1% for placebo, with all events rated as grade 1 or 2.7
In PROSPER, the all-grade AE rates that were higher for enzalutamide versus placebo, respectively, included fatigue (33% vs 14%), hypertension (12% vs 5%), falls (11% vs 4%), and dizziness (10% vs 4%).8 Overall, the rate of serious AEs was 24% with enzalutamide and 18% with placebo, with all events characterized as grade 1 or 2.8
Findings from ARAMIS showed that all-grade AEs were similar for darolutamide versus placebo (83.2% vs 76.9%). The rates of serious AEs of grade 3 or 4 severity were 24.8% for darolutamide and 20.0% for placebo. All-grade AEs that were higher among patients who received darolutamide versus placebo, respectively, included fatigue or asthenic conditions (15.8% vs 11.4%), pain in an extremity (5.8% vs 3.2%), fractures (4.2% vs 3.6%), and rash (2.9% vs 0.9%). As investigators noted, the rates for all-grade AEs were lower among those who took darolutamide compared with those who received placebo, respectively, for falls (4.2% vs 4.7%) and cognitive disorder (0.4% vs 0.2%) and the same for seizures (0.2%).9
“Given the lack of any head-to-head trial, there simply isn’t any way to say which drug is ‘best’ in any meaningful sense,” Fred Saad, MD, an investigator for both the PROSPER and SPARTAN trials, said. “If the results from the different trials were wildly different, or if they had different mechanisms of action, that might suggest differentiation, but they’re not.
“The efficacy data are very similar and impressive, and although there are some differences in reported adverse events, the quality-of-life data are very similar,” said Saad, director of GU Oncology and the Raymond Garneau Chair in Prostate Cancer at the University of Montreal Hospital Centre in Canada. “Although darolutamide did not exclude patients at risk for seizure, the other drugs had extremely low risks of seizures in this patient population. Since patients at high risk for seizures were allowed to be part of ARAMIS but not PROSPER or SPARTAN, it is reasonable that for those patients the first choice would be darolutamide when it becomes available.”