Low-dose Apalutamide in Non-metastatic Castration-resistant Prostate Cancer: A Case Series

Discussion

This study aims to investigate the outcomes of using ≤25% of the recommended apalutamide dose in six patients with non-metastatic castration-resistant prostate cancer. All patients demonstrated significant PSA responses. This finding is consistent with results from an early dose-escalation study conducted over a decade ago, which warranted further investigation but has not been thoroughly explored since.

There is a growing awareness of the general limitation of conventional escalating dose finding. As noted by Brock et al., the majority of dose-escalation clinical trials use methods that seek a maximum tolerated dose (MTD) (9). However, escalation relies on the monotonic efficacy assumption, that higher doses are always more efficacious, which can be violated when the probability of efficacy plateaus at some critical dose (9). In a new guidance entitled Optimizing the Dosage of Human Prescription Drugs and Biological Products for the Treatment of Oncologic Diseases, the US Food and Drug Administration also observed that dose-finding trials for oncology drugs have historically been designed to determine the MTD (10). This MTD paradigm was developed for cytotoxic chemotherapies based on their observed steep dose-response relationships, their limited drug target specificity, and the willingness of patients and providers to accept substantial toxicity due to the lack of effective alternatives for serious life-threatening disease (10). The guidance recognized that modern oncology drugs could demonstrate different dose-response relationships compared to cytotoxic chemotherapy, such that doses below the MTD may have similar activity to the MTD but with fewer toxicities (10). Our study appears in-line with these general observations.

Specifically regarding apalutamide and its MTD, the ARN-509-001 phase 1 study assessed the safety and activity of ARN-509 (apalutamide) in patients with progressive metastatic castration-resistant prostate cancer (11, 12). According to the original registration of that study in 2010, the primary outcome measure was the maximum tolerated dose assessed by dose-limiting toxicities. The secondary outcome measures were: (i) the percent change in PSA relative to baseline and maximal change at 12 weeks, (ii) bone disease assessed by the number of new lesions at 12 weeks, and (iii) soft-tissue disease evaluated by RECIST criteria at 12 weeks (11). The registration update in 2011 replaced “metastatic” with “advanced”, shuffled the outcomes, switched the 50% reduction in PSA at 12 weeks to the primary outcome, and added an assessment of circulating tumor cells (CTCs). The phase 1 study was conducted from July 2010 through May 2012. It enrolled 30 patients who received apalutamide doses from 30 mg to 480 mg (12). A single case presented grade 3 abdominal pain with 300 mg, but 3 other cases of grade 3 adverse events, at undisclosed doses, were considered unrelated to the study treatment. Other than that single case, dose-related registration outcomes were omitted. Table III summarizes the information provided by the report. The study was presented as a single group regardless of dose allocation (Table III, bottom row). Intriguingly, the authors stated that “across all doses tested, antitumor activity was evident as shown by PSA kinetics, radiographic responses, and CTC enumeration” (12). No dose-response relationship was described to clarify the assertion.

The 240-mg dose was determined beforehand as the recommended phase 2 dose (RP2D) when only 24 patients had been enrolled in the dose range of 30-300 mg (13). The choice was not based on any of the predefined outcomes, but on a preclinical assessment (i.e., a mouse xenograft model), on linear and dose-dependent pharmacokinetics, and on imaging showing androgen blockade (13). Twelve patients (55%) had ≥50% PSA decline. Seven patients discontinued the study due to progression. No dose-response efficacy was reported. This study was extended with 2 additional dose levels, 390 and 480 mg, and enrollment totaled 30 patients (14, 15). Eleven of these discontinued the study due to progression. Again, no dose-response efficacy was demonstrated; 42% of patients had ≥ 50% PSA decline. In the final publication (12), the corresponding figures were 16 patients with disease progression and 14 patients (46.7%) with ≥50% PSA decline at 12 weeks. Clearly, increasing the dose neither improved PSA response nor disease control. Pre-registration dose-related data were not mentioned (12). Nevertheless, a poster discussion slide (14) revealed details of the PSA changes as a function of the apalutamide dose.

Figure 2 was reconstructed by digitizing the discussion slide and matching it with the publication waterfall plot of the PSA response at 12 weeks. The plot suggests 2 different groups of patients: low-dose responders, displaying a 66.7% response rate with 60-90 mg doses, and high-dose responders, displaying the same 66.7% response rate with doses ≥300 mg. Due to a lack of individual data, it is not possible to determine which factors contributed to these responses. In any case, the reconstructed waterfall plot of PSA response (Figure 2) emphasizes that the present series of 6 patients responding to ≤60 mg of apalutamide is not a chance finding.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Prostate-specific antigen (PSA) response according to apalutamide dose. The missing apalutamide dose-response waterfall plot. Data extracted by digitizing a conference poster slide (2012 ASCO Annual Meeting, Genitourinary (Prostate) Cancer, Poster Discussion Session) (14), matched with the published plot from (12). Digitizing absolute error estimated at 1%.

In reviewing the published studies, there is no evidence that 240 mg is an optimal dose. Circumstantially, a 240-mg apalutamide dose proved its value against a placebo in the SPARTAN trial (16). The group to which apalutamide was compared were patients with castration-resistant prostate cancer with rising PSA levels that were essentially left untreated. The worst choice of dose (Figure 2 shows that the 240-mg dose had among the lowest response rates) is of course superior to outcomes of the placebo group, as any first-generation anti-androgens also showed superiority when compared to placebo (17-19). Circumstantial data is not evidence, as emphasized in the US Food & Drug Administration’s (FDA’s) Center for Drug Evaluation and Research (CDER) 2018 review (Section 6.3.1: General Pharmacology and Pharmacokinetic Characteristics) (20). The one-line statement about the drug’s minimal effective dose or exposure stated, “Not available”. The CDER review also noted that the maximal tolerated dose or exposure was not reached in the clinical study dosed up to 480 mg once daily. Approval of the drug should not be mistaken as an endorsement or recognition that the 240-mg dose is the best option. Neither minimal effective nor maximal tolerated doses have been identified.

Since no dose response has been assessed, a patient-adapted prescribing approach could be considered instead of the one-size-fits-all single recommended dose. Figure 3 displays an apalutamide prescription flowchart adapted from the observation of patients treated with a similar molecule, enzalutamide (4). The flowchart considers the pattern of apalutamide PSA dose response as shown in Figure 2, the rapid decline of PSA within 2-3 weeks, and the development of adverse events. Starting with a low dose would prevent patients from receiving excessive doses. If responses are observed without adverse events, the low dose should continue (Figure 3, blue path). If severe adverse events are observed, the dose should be reduced until PSA rises again, at which point apalutamide should be discontinued for another therapy (Figure 3, bottom exit path). If no response and no adverse events are observed, the dose should double with PSA monitoring every 3 weeks (Figure 3, red path, right). Thus, the dose increase after 60 mg (time 0) would be 120, 240, and 480 mg at 3, 6, and 9 weeks, respectively, following the same principle of balancing response with adverse events to ensure that the patient receives the appropriate dose rapidly in case a higher dose is needed.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Apalutamide dose prescription flowchart. PSA: Prostate-specific antigen; AE: adverse event.

It is important to note that the flowchart arose from our practice in which most patients with prostate cancer are elderly with several comorbidities. In clinical practice, there are cases that would benefit from higher doses of apalutamide. In the case of a young, fit patient with a life-threatening or symptomatic high disease burden, for whom chemotherapy is not considered or has failed, the flowchart remains applicable without change except for starting with 240 mg instead of 60 mg. In addition, PSA is not the only assessment indicator. At each decision point, clinical evaluation and imaging should be performed, depending on availability and depending on the pattern of practice at the institution.

Low dose apalutamide should not be considered innocuous. Patients in this case series presented some changes over time that appear consistent with known adverse effects (21), including weight loss, stepping difficulties, and hypothyroidism (Table I). One patient presented with steadily decreasing thyroid-stimulating hormone and decreasing thyroxin. The possibility of emerging toxicity from long exposure raises the question of investigating therapeutic pauses or a watchful switch to monotherapy, notably in patients with sustained undetectable PSA levels without evidence of disease progression (22).

Study limitations. First, its retrospective design is subject to selection and recollection bias. Nevertheless, administration of low dose apalutamide was done on consecutive patients without being affected by patients’ characteristics. Except for the oldest patient (95 years old at apalutamide initiation), who was deemed too frail and received an even lower dose (25.7 mg, case #697, Table I), the decision of low dose was not influenced by selection. Second, there was no comparison with a group receiving standard dose, we could not assess if low-dose patients fared worse or better. In addition, imaging procedures and collection of adverse events were not systematically recorded. Lastly, the study included a small sample of African-Caribbean patients and had a short median follow-up of 2.44 years which limited the scope of survival analyses. Mitigating that latter limitation, all patients had a PSA decline by 3 weeks and responded by 3 months. This is an important observation, since early PSA response has repeatedly been shown to be predictive of long-term survival (23, 24). The applicability of low dose apalutamide would require independent prospective confirmation in other populations.