Pathologic Effects of Apalutamide in Lower Risk Prostate Cancer: Results from a Phase II Clinical Trial

Abstract

Introduction

Active surveillance (AS) is a safe and effective strategy for men with lower risk prostate cancer who want to avoid local therapy; however, many patients on AS progress to active treatment (e.g. prostatectomy or radiation). We hypothesized that apalutamide would decrease AS attrition rates through downstaging low-grade tumors.

Methods

This was an open-label, single-arm, Phase II study testing 90 days of oral apalutamide 240 mg daily in men with low to intermediate risk prostate cancer on AS. The primary objective was to determine the percentage of patients with a negative biopsy immediately following treatment. Secondary objectives were to assess long-term clinical outcomes, quality of life (QOL), safety, and biomarkers of response/resistance.

Results

Twenty-three patients enrolled and 22 completed 90 days of apalutamide with post-treatment biopsy. Fifteen (65%) had Grade Group (GG)1 disease and all others had GG2 disease. Seven (30%) had favorable-intermediate risk disease. Of 22 evaluable patients, 13 (59%) had no residual cancer on post-treatment biopsy. The median time to first positive biopsy was 364 days (95% CI: 91–742 days). The impact of apalutamide on QOL was minimal and transient. Decipher risk classifier revealed a greater number of negative post-treatment biopsies in those with higher baseline genomic risk score (P=0.01).

Conclusions

The negative repeat biopsy rate following 90 days of apalutamide was high in men with prostate cancer followed on AS. Apalutamide was safe, well tolerated, and had minimal impact on QOL. Randomized studies evaluating the effects of apalutamide in men enrolled on AS are warranted.

Introduction

Active surveillance (AS) is a strategy aimed at mitigating overtreatment in men with lower risk prostate cancer and entails expectant management in those with low metastatic potential. It is currently endorsed by a number of professional societies and has been increasingly adopted as a standard of care for those with lower risk prostate cancer^1–5. While surveillance strategies for AS vary across centers, this typically involves PSA monitoring, prostate exams and serial prostate biopsies, with referral for local treatment only if pathologic reclassification (i.e. increase in Grade Group [GG] or tumor volume) is observed.^6–8 Multiple series describing the experience with AS have been published, with the data consistently showing that this approach is safe.⁹

AS has been associated with conversion to local treatment in approximately 20–50% of patients^10–13. As such, medical approaches to decrease attrition rates from AS are appealing and the development of potent novel hormonal agents (NHAs) has provided motivation to evaluate their utility in managing men enrolled on AS. Apalutamide, an androgen receptor (AR) inhibitor, is an ideal compound to evaluate in this population given its proven anti-tumor efficacy in men with advanced prostate cancer and the fact that it does not impair the hypothalamic-pituitary-gonadal axis.^{14, 15} Indeed, as a consequence of a negative feedback loop, single-agent therapy with apalutamide results in an increase in testosterone levels. The predicted consequence is that, following treatment discontinuation, the side effects of hypogonadism should dissipate rapidly, leading to minimal disruption in overall quality of life (QOL).

In order to investigate apalutamide as a therapeutic option for prostate cancer patients followed on AS, we conducted a Phase II study to assess the pathologic effects of 3 months of apalutamide on low-to-intermediate risk prostate cancer.

Materials and Methods

Study Design and Patient Population

This was an open-label, single-arm, Phase II study testing 90 days of apalutamide 240 mg daily by mouth in the absence of medical/surgical castration in men followed on AS. Patients were required to have no more than low-to-intermediate risk prostate cancer as determined from prostate biopsy within one year of enrollment, which was defined as: i) clinical stage T1c disease, ii) PSA <15 ng/ml, iii) GG2 present in ≤50% of one core/site, and iv) GG1 disease in all other cores. Following 90 days of treatment, patients underwent repeat prostate biopsy targeting sites of prior positive biopsies (3 from each positive biopsy site) plus 12-core systematic biopsy without targeting. Repeat MRI/TRUS fusion biopsies were performed at the discretion of the treating urologist. Additional prostate biopsies were acquired at 365 and 730 days from the time of enrollment. Otherwise, all patients continued on AS per the standard practice of their treating urologist. This study was approved by our center’s Institutional Review Board and registered with clinicaltrials.gov (NCT02721979).

The primary endpoint was negative repeat biopsy (i.e. absence of residual carcinoma) following 90 days of apalutamide. Secondary endpoints included negative repeat biopsies at Day 365 and Day 730 following enrollment, exit from AS due to pathologic progression or otherwise at 2 years, receipt of local treatment for prostate cancer, QOL changes, and safety (as assessed by the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.03).¹⁶ QOL was assessed using the FACT-P and SF-36 surveys. Exploratory transcriptional analyses were conducted on formalin fixed paraffin embedded tissue from men enrolled to this study. Tumor expression profiles were generated with a clinical-grade transcriptome assay (Veracyte Inc. San Diego) utilizing the Human Exon 1.0 ST oligonucleotide microarray (ThermoFisher, Carlsbad CA). At least 0.5 mm of tumor sample in a biopsy was required. Details on the statistical plan can be found in the Supplemental Methods and Data.

Results

Treatment Effects

Between October 2017 and December 2019, 23 patients enrolled on this trial. Table 1 presents demographic data. The median time on AS was 10.4 months (IQR: 3.7–40.1 months) at the time of enrollment and 13/23 (57%) subjects had ≥2 prior positive biopsies while on AS. One patient had a single negative biopsy prior to enrolling on study (5/6 prior biopsies were positive). All subjects completed the 90-day treatment course, and 22 underwent post-treatment biopsy, with one patient dropping out secondary to the SARS-COV2 pandemic and a desire to limit clinic exposure. The study was stopped prior to enrolling the planned 33 patients due to slow enrollment.

Table 1:

Patient demographics at baseline.

Characteristic	N = 23
Age at diagnosis, median (range)	64 (45–72)
Age at enrollment, median (range)	67 (45–76)
Year of diagnosis, N (%)
2010–2014	5 (22%)
2015–2019	18 (78%)
PSA, median (range)	5.1 (2.1–12.3)
Grade group, N (%)
1	15 (65%)
2	8 (35%)
Number of cores involved, median (range)	2 (1–6)
NCCN risk category, N (%)
Very low	6 (26%)
Low	10 (43%)
Favorable intermediate	7 (30%)

Of 22 evaluable patients, 13 (59%; 95% CI 36%−79%; P<0.001 compared to 20% null rate based on an exact binomial test) had no evidence of residual cancer on post-treatment biopsy (i.e. Day 91). Assuming we enrolled the entire 33 patient cohort and all additional patients had residual cancer post-treatment, the observed negative biopsy rate would have been 13/33 (39%; 95% CI 23%−54%; P=0.008), which was still sufficient to reject the null hypothesized response rate of ≤20%. Only one subject had an MRI/TRUS fusion biopsy on Day 91, which was positive for residual cancer.

Twenty-one subjects underwent biopsy at Day 365 and 19 at Day 730. Of note, one patient refused the Day 365 biopsy but subsequently did undergo the Day 730 biopsy. Seven of 21 (33%) patients had no evidence of residual cancer on the Day 365 biopsy. Four of 19 (21%) had a negative biopsy at Day 730. Of note, one subject had a positive biopsy at Day 365 and a negative Day 730 biopsy. Median time to first positive biopsy was 364 days (95% CI: 91 – 742 days) (Figure 1).

Figure 1:

Positive biopsy-free survival with 95% confidence limits (shaded gray areas).

At Day 91, all patients had >50% decline in PSA while on apalutamide and 15 (65%) had a ≥90% PSA decline (Figure 2A). Following cessation of apalutamide, PSA rose in all patients and by Day 365 the median PSA was not significantly different compared to baseline (median difference −0.2 ng/mL, 95% CI: −0.9 to 0.5, P=0.5) (Figure 2B, Table S1). During the treatment period (i.e. through Day 91), median testosterone levels were significantly elevated compared to baseline (median difference 275, 95% CI: 216 to 340, P<0.001) (Figure 2C, Table S2).

Figure 2:

On study changes in PSA and testosterone. A) Waterfall plot depicting largest percent change in PSA relative to baseline, B) PSA in ng/mL, and C) testosterone in ng/dL over the course of the study (median values with IQR). *P-value ≤0.05 compared to Day 1 (Wilcoxon signed rank tests).

Over a median follow-up of 753 days (IQR: 747–762 days), 5 patients underwent definitive local therapy. Four received local therapy due to GG reclassification (3 received radiation therapy; 1 underwent radical prostatectomy) and one subject opted for radical prostatectomy who had stable GG2 disease compared to baseline. Local therapy was received between 464 to 918 days after enrolling on study. Two additional patients (6 total) had an increase in GG from pre-study baseline while on trial. One was lost to follow-up. The other had an increase to GG2 disease and opted for ongoing AS. Of note, subsequent biopsy showed a decline back to GG1 and Prolaris testing (Myriad Genetics) indicated low risk of death (2.3% at 10 years) with AS.^17–19 Two other patients dropped out prior to Day 730: one died of an unrelated cause and the other developed a second metastatic primary.

Safety and Quality of Life Effects

Apalutamide was well tolerated and adverse events were generally consistent with apalutamide’s known safety profile (Table 2). One patient had grade 3 hypertension and another had grade 3 rash. Both of these patients were able to remain on study following dose reductions.

Table 2:

Treatment-related adverse events in ≥5% of subjects. Data is provided for subjects who received at least one dose of study drug (N=23).

Adverse Event	Grade 1	Grade 2
Fatigue	16 (70%)	2 (9%)
Gynecomastia	16 (70%)
Arthralgia/myalgia	7 (30%)
Dysgeusia	7 (30%)
Rash	6 (26%)
Cognitive impairment	5 (22%)
Hot flashes	5 (22%)
Elevated TSH	4 (17%)
Anorexia	3 (13%)
Dry skin	3 (13%)
Libido decreased	3 (13%)
Pruritus	3 (13%)
Nausea	2 (9%)
Weight loss	2 (9%)

Transient declines across several QOL scores were noted from Day 1 to Day 91; however, all median summary scores were near baseline by Day 180 (Figure 3). It is notable that the median changes in QOL scores were not considered clinically meaningful, with the exception being the SF-36 energy/fatigue score, which fell by a median of 13 points (95% CI 5 to 25; P=0.01) from Day 1 to Day 91^20-22. Interestingly, several participants reported improvements in QOL beyond Day 91 as compared to baseline (Figure 3 and Table S3).

Figure 3:

Summary quality of life changes over time. A) FACT-P summary scores, and B) SF-36 summary scores (median values with IQR). *P-value ≤0.05 compared to Day 1 (Wilcoxon signed rank tests).

Transcriptional Assessments

There were no significant associations between responses and GG (P=0.6) or NCCN risk category (P=0.3). Responders had higher Decipher risk scores scores at baseline (P=0.01) and there was marginal-to-weak evidence of enrichment for higher Cuzick cell cycle progression risk score in responders (P=0.14) (Figure 4)^17–19. These signatures are prognostic in various setting of prostate cancer, with higher scores correlating with more aggressive disease^{17–19, 23}. Additional details on transcriptional profiling studies can be found in the Supplemental Methods and Data (Figure S1-S4).

Figure 4:

Transcriptional risk profiling. Response (i.e., negative Day 91 biopsy) stratified by baseline Decipher risk score (Left) and Cuzick risk score (Right).

Discussion

We observed a high negative biopsy rate immediately following 3 months of apalutamide monotherapy in men followed on AS. Side effects of apalutamide in this patient population were minimal, and therapy was generally well tolerated. Importantly, we did not observe any long-term negative impact on QOL. This may be due to apalutamide’s short half-life and the compensatory increase in circulating androgen levels that occur as a consequence of AR antagonism.

It is also notable that we enrolled a patient population that was higher risk than prior therapeutic AS studies, with 30% demonstrating favorable intermediate risk disease^{24, 25}. While correlative studies were exploratory in nature, we did observe that those with higher genomic risk scores responded favorably to treatment. Given that this molecularly defined subgroup may be more likely to demonstrate adverse pathologic features at the time of prostatectomy, the use of NHAs in those with higher risk prostate cancer as determined using novel genomic risk stratifiers should be further validated²⁶.

While this was a non-randomized study, it is worth noting that the observed negative repeat biopsy rate of 59% immediately following treatment compares quite favorably to historic controls. The randomized Phase II REDEEM study demonstrated a negative repeat biopsy rate of 28% following 18 months of dutasteride²⁴. A second trial by Cussenot and colleagues tested a single SC injection of leuprolide 22.5 mg plus 15 days of oral bicalutamide and reported a negative repeat biopsy rate of 45% approximately one year after enrollment.²⁵ Finally, the Phase II ENACT study randomized patients with low or intermediate risk prostate cancer to enzalutamide for one year (N=114) vs. AS alone (N=113)²⁷. Time to pathologic or therapeutic progression was found to be significantly reduced with enzalutamide and the negative biopsy rate was 43% (40 of 53 patients) at one year, although there were 21 subjects with missing biopsy results at that timepoint. The negative biopsy rate at two years was 29% among those with biopsy results available (N=66).

Prior editorials have pointed out that medical interventions in patients enrolled to AS are incongruent with the goal of avoiding treatment.^{28, 29} However, it is hard to ignore the fact that many men on AS still receive local therapy – often on the recommendation of their urologist following pathologic reclassification. Because local treatment can result in long-term erectile and urinary dysfunction, the development of non-invasive approaches aimed at allowing men to avoid surgery and radiation are still justified. In order to balance these competing factors, medical interventions should be reserved for those at highest risk of disease progression (e.g. high genomic risk and/or ≥GG2 disease) and aim to minimize drug exposure. We observed that apalutamide when given for 3 months had an acceptable safety profile and minimal impact on QOL. Furthermore, we observed that short course apalutamide had similar clinical effects to a full year of enzalutamide. As such, larger studies evaluating short course NHA therapy should be prioritized in this population²⁷.

This study was designed to evaluate for preliminary evidence that apalutamide was effective in men enrolled onto AS. As such, pathologic assessment from post-treatment biopsy was selected as an early indicator of efficacy. While we acknowledge that a negative biopsy is not a validated proxy for long-term outcomes, this was felt to be a reasonable primary endpoint given that the majority of AS programs rely on pathologic changes as a trigger to recommend definitive local treatment^6–8. We acknowledge that the small sample size of this study is one of its primary limitations, which was further challenged by its premature termination. However, in spite of this shortcoming, we still detected a robust efficacy signal, and future prospective studies of apalutamide in this patient population are warranted.

Ultimately, large randomized studies will be needed to evaluate if systemic therapies are useful in the management of patients with prostate cancer followed on AS. However, the design of such a trial is hampered by the lack of endpoints that would clearly establish clinical benefit. Because metastases and death rarely occur in men followed on AS, novel endpoints need to be utilized in this patient population⁹. Receipt of definitive local therapy appears an obvious candidate for a primary endpoint given that mitigating overtreatment is one of the primary goals of AS; however, the reasons for pursuing local treatment are varied and often not related to objective changes in prostate cancer risk. Indeed, one large multicenter study reported that 13% of patients on AS underwent local treatment in the absence of disease progression³⁰. It is plausible that some of these patients opted for local treatment solely based on PSA changes, which would further complicate the conduct of a randomized, placebo-controlled study given that NHAs will suppress PSA and therefore make blinding challenging. One option to overcome this issue would be to create a hybrid endpoint, whereby the receipt of definitive therapy is only considered clinically meaningful if occurring as a result of pathologic reclassification.

Conclusions

In spite of this study’s small sample size, we observed a high negative repeat biopsy rate in prostate cancer patients followed on AS. Of note, not only did low risk patients appear to benefit but we also saw favorable pathologic effects in those with higher risk features (i.e. GG 2 disease, high genomic risk) – indicating that men at risk for needing local treatment could benefit from this approach. Based on these findings, future prospective studies of apalutamide in men with prostate cancer followed on AS are justified.

Abbreviations

AS

active surveillance

QOL

quality of life

GG

grade group

NHA

novel hormonal agent

AR

androgen receptor

PSA

prostate specific antigen

FACT-P

functional assessment of cancer therapy – prostate

FACT-G

functional assessment of cancer therapy – general

SF-36

36-item short form

NCCN

National Comprehensive Cancer Network