FDA Approval Summary: Lutetium Lu 177 Vipivotide Tetraxetan for Patients with Metastatic Castration-Resistant Prostate Cancer

Patients with metastatic castration-resistant prostate cancer (mCRPC) whose disease progresses after androgen receptor pathway inhibitors (ARPI) and taxanes have limited treatment options. In addition, a subset of patients may not be medically fit to receive some available therapies due to the risk of severe toxicities. On March 23, 2022, the FDA approved Pluvicto (lutetium Lu 177 vipivotide tetraxetan, also known as ¹⁷⁷Lu-PSMA-617) for the treatment of adult patients with prostate-specific membrane antigen (PSMA)-positive mCRPC who have been treated with AR pathway inhibition and taxane-based chemotherapy (1). The FDA granted approval based on a statistically significant and clinically meaningful improvement in overall survival (OS) when ¹⁷⁷Lu-PSMA-617 was combined with best standard of care (BSoC) compared with BSoC alone in a large, adequate, and well-controlled randomized phase III trial. On the same day, the FDA approved Locametz (kit for the preparation of gallium Ga 68 gozetotide injection, also known as ⁶⁸Ga-PSMA-11), a radioactive diagnostic agent indicated for PET of PSMA-positive lesions, for selection of patients with metastatic prostate cancer for whom ¹⁷⁷Lu-PSMA-617 PSMA-directed therapy is indicated (2–4). This article summarizes the data and the FDA perspective supporting traditional approval of ¹⁷⁷Lu-PSMA-617 for this indication.

PSMA is a transmembrane protein which has high expression in prostatic epithelium. PSMA expression is substantially higher in malignant prostate cancer cells (5). ¹⁷⁷Lu-PSMA-617 is a radioligand therapeutic agent consisting of radionuclide lutetium-177 linked to a peptide moiety (vipivotide tetraxetan also known as PSMA-617). PSMA‐617 is a small-molecule inhibitor which binds with high affinity to the extracellular domain of PSMA. ¹⁷⁷Lu-PSMA-617 delivers radiation to PSMA-expressing prostate cancer cells via its binding to PSMA-expressing cells, resulting in subsequent cell death by beta emission from lutetium-177 (6).

In addition to prostate cancer cells, PSMA is expressed on epithelium of benign tissues including kidneys, salivary glands, lacrimal glands, liver, spleen, and in small and large bowel (6, 7). Expected toxicities of ¹⁷⁷Lu-PSMA-617 are associated with radiation exposure to the PSMA-expressing tissues, surrounding tissues and organs adjacent to the PSMA-expressing tumors, and the organs involved in urinary excretion of ¹⁷⁷Lu-PSMA-617. As a radioactive product, ¹⁷⁷Lu-PSMA-617 is genotoxic and carcinogenic, has the potential to cause embryo-fetal toxicity and impair human fertility.

The recommended ¹⁷⁷Lu-PSMA-617 dosing regimen of 7.4 GBq (or 200 mCi) administered intravenously once every 6 weeks for a total of up to six doses (cumulative dose = 44.4 GBq) was based on the demonstration of favorable efficacy and safety in the VISION study.

Pharmacokinetics, dosimetry, biodistribution, and QT prolongation assessments of ¹⁷⁷Lu-PSMA-617 were conducted in a limited number of patients (n = 30) in a VISION substudy.

¹⁷⁷Lu-PSMA-617 is eliminated predominantly through renal excretion. Limited dosimetry data in the VISION substudy indicated a trend for increased kidney radiation exposure with decreasing creatinine clearance (CLcr), with 2-fold higher kidney radiation in the only patient enrolled with CLcr < 60 mL/minute (54 mL/minute) compared with patients with normal renal function (CLcr: ≥90 mL/minute). Also, the analysis suggested the potential for cumulative kidney radiation exposure to reach or exceed the kidney radiation safety threshold (23 Gy based on external beam radiotherapy) after the fifth or sixth doses of ¹⁷⁷Lu-PSMA-617 in patients with mild (CLcr: 89 to 60 mL/minute) and moderate (CLcr: 59 to 30 mL/minute) renal impairment. In the VISION study, among patients who received ¹⁷⁷Lu-PSMA-617, higher incidence of toxicity and dose modifications and discontinuations due to adverse reactions were observed in patients with moderate renal impairment (n = 58) compared with patients with mild renal impairment (n = 173) or normal renal function (n = 283). Among patients who received Pluvicto, the frequency of grade 3–4 renal toxicity-related adverse reactions was 1.4% in patients with normal renal function, and 2.9%, and 14% in patients with mild and moderate renal impairment, respectively. Consequently, frequent monitoring of adverse reactions was recommended for patients with moderate renal impairment receiving ¹⁷⁷Lu-PSMA-617. No patients with severe renal impairment (CLcr: 29 to 15 mL/minute) were enrolled in the VISION study.

The efficacy of ¹⁷⁷Lu-PSMA-617 was evaluated in VISION (NCT03511664), a randomized (2:1), multicenter, open-label trial that evaluated ¹⁷⁷Lu-PSMA-617 plus BSoC or BSoC alone in men with progressive, PSMA-positive mCRPC. Patients across North America (Canada, Puerto Rico, and United States) and Europe (Belgium, Denmark, France, Netherlands, Sweden, and United Kingdom) were screened for study entry. Randomization was stratified by baseline lactase dehydrogenase (LDH), presence of liver metastases, Eastern Cooperative Oncology Group performance status (ECOG PS) score, and inclusion of an ARPI as part of BSoC at the time of randomization. All patients received a gonadotropin releasing hormone analog or had prior bilateral orchiectomy. Patients were required to have received at least one ARPI, and one or two prior taxane-based chemotherapy regimens. A taxane regimen was defined as a minimum exposure of two cycles of a taxane-based chemotherapy. If a patient had received only one taxane regimen, the patient was eligible if the patient's physician deemed that patient unsuitable to receive a second taxane regimen (e.g., frailty assessed by geriatric or health status evaluation or intolerance, etc.). Eligible patients were required to have PSMA-positive mCRPC defined as having at least one tumor lesion with ⁶⁸Ga-PSMA-11 uptake greater than normal liver. Patients were excluded if any lesions exceeding size criteria in short axis [organs ≥ 1 cm, lymph nodes ≥ 2.5 cm, bones (soft-tissue component) ≥ 1 cm] had uptake less than or equal to uptake in normal liver.

Patients received ¹⁷⁷Lu-PSMA-617, 7.4 GBq (200 mCi) every 6 weeks for up to a total of six doses plus BSoC or BSoC alone. BSoC administered at the investigator's discretion included ketoconazole, radiotherapy to localized prostate cancer targets, bone-targeted agents, androgen-reducing agents, and ARPIs. Investigational agents, cytotoxic chemotherapy (e.g., cabazitaxel), immunotherapy, other systemic radioisotopes (e.g., radium-223), or hemibody radiotherapy treatment were not allowed as part of BSoC. Patients continued treatment for up to six doses, or until disease progression or unacceptable toxicity. Patients with stable disease or a partial response after four doses of ¹⁷⁷Lu-PSMA-617 received up to two additional doses per investigator's discretion.

The co-primary endpoints were OS and radiographic progression-free survival (rPFS) by blinded independent central review (BICR) per Prostate Cancer Working Group 3 criteria (8). The study was considered positive if either OS or rPFS (or both) was statistically significant. The key secondary endpoints were overall response rate (ORR) by BICR per RECIST v1.1 (9), and time to symptomatic skeletal events (SSE), which was defined as the time from randomization to the first new symptomatic pathologic bone fracture, spinal cord compression, tumor-related orthopedic surgical intervention, requirement for radiotherapy to relieve bone pain, or death from any cause, whichever occurred first. The design provided at least 90% power for OS and 84% for rPFS, with an overall one-sided type I error rate ≤ 0.025. The key secondary endpoints were controlled for multiplicity using the Hochberg closed test procedure and used the alpha level from the successful OS results for testing. An OS interim analysis was planned to be performed at the time of the primary rPFS analysis.

Early in trial conduct, a substantial and disproportionate rate of dropout (mainly due to withdrawal of consent) was seen in the control arm (56% in the first 84 patients) compared with the investigational arm (1.2% in the first 166 patients). To reduce the dropout rate and its effect on trial outcomes, the sponsor implemented mitigation strategies including enrollment caps for study sites with a high rate of dropout, and enhanced education and interdisciplinary communication at study sites. These mitigation strategies reduced subsequent drop out; however, the rate of drop out remained higher among patients enrolled to the control arm compared with the investigational arm after implementation (16.3% in 196 patients and 4.2% in 385 patients, respectively). In addition, the target sample size was increased from N = 750 to N = 814 to maintain statistical power, and a portion of OS events for those who withdrew consent and were lost to follow-up were ascertained through available registries [34 (41%) of 82 patients who withdrew consent were later reported as dead through public registry search]. The OS analysis was performed in all patients who were randomized to a treatment arm (data cut-off date: January 27, 2021). Because rPFS or SSE could not be obtained after withdrawal of consent, rPFS and SSE analyses were only performed in the population randomized on or after March 5, 2019, when mitigation strategies were implemented to reduce the effect of asymmetric dropout.

Key demographics and baseline disease characteristics of patients enrolled in VISION are summarized in Table 1: 551 patients were randomized to the ¹⁷⁷Lu-PSMA-617 plus BSoC arm, while 280 patients were randomized to the BSoC alone arm. All patients had received at least one prior taxane-based chemotherapy regimen and 41% of patients received two. One prior ARPI had been administered to 51% of patients, 41% of patients received two, and 8% of patients received three or more. During the treatment period, 53% of patients in the ¹⁷⁷Lu-PSMA-617 plus BSoC arm and 68% of patients in the BSoC alone arm received at least one ARPI. Overall, the demographic and disease characteristics were well balanced between the treatment arms (Table 1).

The co-primary endpoint of OS was statistically significant, with patients randomized to receive ¹⁷⁷Lu-PSMA-617 plus BSoC having prolonged OS (median estimate 15.3 months) compared with BSoC alone (median estimate 11.3 months), HR, 0.62 [95% confidence interval (CI): 0.52–0.74; P < 0.001] (Table 2; Fig. 1). The planned interim OS analysis was not necessary as the targeted number of OS events were observed before the targeted number of rPFS events. Therefore, the final OS analysis was evaluated at the time of rPFS analysis.

Where feasible, survival status for patients who dropped out was obtained via public registries and this was specified in the site-specific informed consent forms. Thus, asymmetric censoring due to withdrawal of consent was reduced for OS events in the primary analysis. Nonetheless, in the OS analysis set, 15 patients (2.7%) were censored because of withdrawal of consent in the experimental arm compared with 22 patients (11.8%) in the control arm. Conservative sensitivity analyses were conducted by both the Sponsor and the FDA considering extreme scenarios and the potential for informative censoring due to disproportionate drop out. The Sponsor conducted (i) an analysis which considered all dropouts in the ¹⁷⁷Lu-PSMA-617 plus BSoC arm as events, (ii) two analyses where data were imputed for dropouts in the control arm based on the HR in the 20% of patients with the longest survival either overall or in the BSoC arm, and (iii) a tipping-point analysis which quantified the increase or decrease in the risk of events in patients dropping out of the ¹⁷⁷Lu-PSMA-617 plus BSoC arm or the BSoC arm that would make the primary analysis lose statistical significance. These analyses supported the primary OS findings. The FDA conducted an additional OS sensitivity analysis that excluded patients who withdrew consent. The HR was 0.62 (95% CI: 0.52–0.75), which was consistent with the primary OS analysis. The results of the sensitivity analyses of OS supported the statistically robust finding of superiority for OS results.

The co-primary endpoint of BICR assessed rPFS was statistically significant, with patients randomized to receive ¹⁷⁷Lu-PSMA-617 plus BSoC having prolonged rPFS (median estimate 8.7 months) compared with BSoC alone (median estimate 3.4 months), HR, 0.40 (95% CI: 0.31–0.52; P < 0.001). As described previously, the PFS analysis set was a subset of the overall intention-to-treat (ITT) analysis set and included 385 patients in the ¹⁷⁷Lu-PSMA-617 plus BSoC arm and 196 patients in the BSoC alone arm who were enrolled after mitigation strategies were implemented to reduce early withdrawal of consent. However, there remained approximately 20% more censoring of rPFS in the control arm compared with the investigational arm and adequate tumor assessments were not available for 18% of the patients in the control arm compared with only 1.3% on the ¹⁷⁷Lu-PSMA-617 arm. While sensitivity analyses of the rPFS results due to early withdrawal were supportive of a statistically superior effect favoring the ¹⁷⁷Lu-PSMA-617 arm, interpretation of the magnitude of the rPFS effect was limited due to the highly disproportionate degree of censoring between the treatment arms.

Key secondary endpoints were supportive of the primary efficacy results. The secondary endpoint ORR by RECIST v1.1 was evaluated in the 319 patients in the experimental arm and 120 patients in the control arm who had evaluable disease at baseline. Of these, 95 (30%) in the experimental arm and 2 (2%) in the control arm had an objective response per RECIST (v1.1; Table 2).

A delay in the time to SSEs was demonstrated favoring the ¹⁷⁷Lu-PSMA-617 arm. Of note, death was included as an event in the SSE endpoint definition, and most of the events in the SSE analysis were death events. Of 256 SSE (66.5%) in the ¹⁷⁷Lu-PSMA-617 arm and 137 SSE (70%) in the BSoC arm, only 60 events (15.6%) and 34 events (17.3%), respectively, were non-death events.

Of the 831 patients randomized, 734 patients received at least one dose of randomized treatment (¹⁷⁷Lu-PSMA-617 plus BSoC: n = 529, and BSoC alone: n = 205). The median duration of exposure to ¹⁷⁷Lu-PSMA-617 plus BSoC was 7.8 months. Among patients who received ¹⁷⁷Lu-PSMA-617 plus BSoC, the median number of doses of ¹⁷⁷Lu-PSMA-617 received was 5 (range, 1–6). The median cumulative dose of ¹⁷⁷Lu-PSMA-617 was 37.5 GBq (range, 7.0–48.3). The median duration of follow-up was 14.8 months for patients receiving ¹⁷⁷Lu-PSMA-617 plus BSoC.

Serious adverse reactions occurred in 36% of patients who received ¹⁷⁷Lu-PSMA-617 plus BSoC. Fatal adverse reactions occurred in 2.8% of patients who received ¹⁷⁷Lu-PSMA-617 plus BSoC, including sepsis (0.9%), pancytopenia (0.6%), hepatic failure (0.4%), intracranial hemorrhage (0.2%), subdural hematoma (0.2%), ischemic stroke (0.2%), COVID-19 (0.2%), and aspiration pneumonia (0.2%). The two fatal adverse reactions of intracranial hemorrhage and subdural hematoma occurred in patients who had concurrent treatment-related thrombocytopenia.

The most common adverse reactions (≥20%) occurring at a higher incidence in patients who received ¹⁷⁷Lu-PSMA-617 plus BSoC were fatigue, dry mouth, nausea, anemia, decreased appetite, and constipation. The most common laboratory abnormalities that worsened from baseline in ≥ 30% of patients who received ¹⁷⁷Lu-PSMA-617 plus BSoC were decreased lymphocytes, decreased hemoglobin, decreased leukocytes, decreased platelets, decreased calcium, and decreased sodium.

The approval decision was based on a statistically significant and clinically meaningful improvement in OS demonstrated by adding ¹⁷⁷Lu-PSMA-617 to BSoC. Efficacy results were considered in the context of an acceptable safety profile in a patient population with unmet medical need. The FDA focused its assessment of efficacy on the OS benefit, using rPFS in the modified analysis population as well as durable ORR results from the investigational arm as supportive evidence. In addition to the strength of OS as a clinically meaningful endpoint, loss to follow-up of patients who withdrew consent was mitigated for the OS endpoint by ascertainment of survival status for many patients via public registries. Assessment of OS events for these patients enabled the review team to reliably assess OS in the ITT population. The OS benefit was robust and held up to a series of strict sensitivity analyses that assessed the impact of remaining asymmetric censoring due to drop out for those patients whose OS status could not be obtained.

In contrast to standard RECIST PFS, radiographic PFS in prostate cancer relies heavily on bone scan findings which have less resolution than CT scans and require a more complicated algorithm to define a progression event (10). In addition, the meaningfulness of a tumor progression endpoint relies heavily on a large magnitude of effect in the context of the natural history of the disease and safety findings, among other factors (10, 11). In VISION, there was a substantially higher proportion of rPFS censoring in the control arm. Baseline characteristics of patients who withdrew consent were generally balanced and did not demonstrate clear evidence of informative censoring, and FDA conducted strict sensitivity analyses to ensure that the asymmetric censoring did not have a substantial impact on the conclusion of rPFS superiority. However, asymmetric censoring of patients between treatment arms can lead to overestimation or underestimation of the true magnitude of effect, which is critical to interpreting the clinical relevance of a result. While sensitivity analyses supported the statistical difference in rPFS favoring ¹⁷⁷Lu-PSMA-617, asymmetric censoring created uncertainty around the true magnitude of delay in tumor progression, and as such, numeric results for rPFS were not included in FDA product labeling.

The critical review issue for this application was the highly asymmetric dropout. A potential explanation for the higher rate of dropout in the control arm may be lack of patient- and/or investigator-perceived equipoise at randomization given the exclusion of FDA-approved therapies such as cabazitaxel from the control arm, as well as the public information on the potential efficacy and safety of 177Lu-PSMA-617. Despite the trial eligibility criteria mandating exclusion of patients who may have been suitable for treatment with a second taxane, 135 (16%) of 831 patients went on to receive cabazitaxel as a subsequent therapy, suggesting that a more active control arm may have been feasible. While FDA does not have a comparative efficacy standard for traditional drug approval and thus cannot mandate a specific control arm, a trial design that may be perceived as lacking equipoise due to the choice of comparator may lead to asymmetric withdrawal and render the study results difficult to interpret.

This study was unusual in that the 508 OS events intended for the final OS analysis were reached before the prespecified 364 rPFS events required for the primary rPFS analysis. This was due to a larger number of patients in the OS analysis set compared with the modified rPFS analysis set. Furthermore, patients enrolled in both arms died sooner than anticipated. The speed at which the OS events accumulated resulted in the final OS analysis coinciding with the time of the interim OS analysis per protocol which was planned to occur at the time of rPFS analysis.

The SSE endpoint included death as an event which is different than the SSE definition used to support a prior FDA approval in mCRPC (12). The difference in HR for SSE was driven by death events between the arms rather than by the other skeletal-related events outlined in the definition of an SSE, which limited interpretation of the clinical utility of this endpoint. In addition, sensitivity analyses to take into account asymmetric censoring were not applied to the SSE endpoint. Because of these issues, time to SSE was not included in product labeling.

In VISION, patients were excluded if any lesion exceeding size criteria in short axis had uptake less than or equal to uptake in normal liver. FDA's review of a VISION ⁶⁸Ga-PSMA-11 reader substudy found that PET interpretation to select patients for ¹⁷⁷Lu-PSMA-617 may differ depending on imaging readers (i.e., high inter-rater variability). This finding supports the need for additional investigation aimed at optimizing ⁶⁸Ga-PSMA-11 PET imaging interpretation criteria for ¹⁷⁷Lu-PSMA-617 patient selection. Toward this end, FDA issued a postmarketing commitment to study outcomes specific to patients treated under expanded imaging eligibility criteria.

Black or African American patients were underrepresented in VISION, with only 7% accrual in the setting of an approximately 14% Black or African American U.S. population. This is particularly notable given U.S. Black men are more likely to be diagnosed with prostate cancer and have higher prostate cancer mortality compared with non-Hispanic White men (13). Patients from racial and ethnic minorities should be represented appropriately in clinical trials to reflect the diversity of the U.S. population that will ultimately use the treatment. The FDA Oncology Center of Excellence has initiated several programs to encourage sponsors to submit diversity plans that include a multipronged approach to ensure that clinical trial sites recruit a diverse study population (14, 15).

Exposure to radiotherapeutics has been associated with delayed toxicities such as secondary malignancies, including myelodysplastic syndrome and acute myeloid leukemia (16). The median duration of follow-up at the time of data cutoff was 14.8 months for patients receiving ¹⁷⁷Lu-PSMA-617, which was not sufficient to define the magnitude of the risk for key radiation-induced delayed toxicities. FDA issued a postmarketing requirement (PMR) to conduct an integrated analysis of the safety outcomes from the ongoing clinical trials of ¹⁷⁷Lu-PSMA-617 [VISION study and its substudy, Trial CAAA617C12301 (NCT04720157), Trial CAAA617B12302 (NCT04689828), and other clinical trials as appropriate)] to further characterize the long-term outcomes of the known serious toxicities and potential clinically important delayed toxicities of ¹⁷⁷Lu-PSMA-617. In addition, due to the concern for high radiation exposure to the kidneys, potential for long-term radiation toxicity, and paucity of dosimetry data in patients with moderate and severe renal impairment, FDA issued a PMR to conduct a clinical trial to determine the kidney dosimetry and long-term safety of ¹⁷⁷Lu-PSMA-617, and potential for dose adjustment of ¹⁷⁷Lu-PSMA-617 in these patients. To reduce the risk of renal toxicity, patients should be advised to remain well hydrated and to urinate frequently before and after administration of Pluvicto.

On March 23, 2022, the FDA approved ¹⁷⁷Lu-PSMA-617 for the treatment of adult patients with PSMA-positive mCRPC who have been treated with androgen receptor pathway inhibition and taxane-based chemotherapy. The approval was based on a meaningful advantage in OS supported by a delay in tumor progression and higher objective response rate. The decision took into account several important factors unique to this application's context (Table 3). mCRPC is a life-threatening condition that lacks curative treatment, creating an unmet need. This product provides a novel mechanism of action as the first radioligand therapeutic agent for mCRPC providing antitumor effects in both bone and non-bone sites and may provide an opportunity to be combined with other available therapies with different toxicity profiles across mechanistic classes. Finally, ¹⁷⁷Lu-PSMA-617 offers a different safety profile than other available systemic therapies which was determined to be acceptable for the indicated patient population, expanding treatment options that can be individualized to a patient's preferences and comorbidities.

No disclosures were reported.

The Editor handling the peer review and decision-making process for this article has no relevant employment associations to disclose.