Visual Abstract
Abstract
177Lu-labeled prostate-specific membrane antigen (PSMA) radioligand therapy (RLT) is a new treatment option for metastatic castration-resistant prostate cancer (mCRPC). Its low toxicity profile favors use in elderly patients or in patients with critical comorbidities. The purpose of this analysis was to evaluate the efficacy and safety of [177Lu]-PSMA RLT in mCRPC patients at least 80 y old. Methods: Eighty mCRPC patients at least 80 y old underwent [177Lu]-PSMA-I&T RLT and were retrospectively selected. The patients were previously treated by androgen receptor–directed therapy, received taxane-based chemotherapy, or were chemotherapy-ineligible. The best prostate-specific antigen (PSA) response was calculated, as well as clinical progression-free survival (cPFS) and overall survival (OS). Toxicity data were acquired until 6 mo after the last treatment cycle. Results: Of 80 patients, 49 (61.3%) were chemotherapy-naïve and 16 (20%) had visceral metastases. The median number of previous mCRPC treatment regimens was 2. In total, 324 cycles (median, 4 cycles; range, 1–12) with a median cumulative activity of 23.8 GBq (interquartile range, 14.8–42.2) were applied. A PSA decline of 50% was achieved in 37 (46.3%) patients. Chemotherapy-naïve patients showed higher 50% PSA response rates than chemotherapy-pretreated patients (51.0% vs. 38.7%, respectively). Overall, median cPFS and OS were 8.7 and 16.1 mo, respectively. The median cPFS and OS of chemotherapy-naïve patients were significantly longer than those of chemotherapy-pretreated patients (10.5 vs. 6.5 mo and 20.7 vs. 11.8 mo, respectively, P < 0.05). A lower hemoglobin level and higher lactate dehydrogenase level at baseline were independent predictors of shorter cPFS and OS. Treatment-emergent grade 3 toxicities were anemia in 4 patients (5%), thrombocytopenia in 3 patients (3.8%), and renal impairment in 4 patients (5%). No nonhematologic grade 3 and no grade 4 toxicities were observed. The most frequent clinical side effects were grade 1–2 xerostomia, fatigue, and inappetence. Conclusion: [177Lu]-PSMA-I&T RLT in mCRPC patients at least 80 y old is safe and effective, comparable to previously published data on non–age-selected cohorts with a low rate of high-grade toxicities. Chemotherapy-naïve patients showed a better and longer response to therapy than taxane-pretreated patients. [177Lu]-PSMA RLT seems to be a meaningful treatment option for older patients.
- metastatic castration-resistant prostate cancer
- radioligand therapy
- prostate-specific membrane antigen
- [177Lu]-PSMA
- elderly patients
Octogenarians represent a considerable proportion of patients with prostate cancer (1). Future projections indicate a growth due to the global increase in life expectancy and more life-prolonging treatments (2). Management of late-stage disease in elderly men is challenging given the reduced normal organ function or higher frequency of drug interactions related to treatment of comorbidities (3). Elderly patients have higher incidences of cardiovascular diseases, impaired renal function, and a lower bone marrow function. Therapy-related side effects are often less tolerated by senescent patients (4). In some age-stratified analyses of cytotoxic and androgen receptor–targeted agents in older patients with metastatic castration-resistant prostate cancer (mCRPC), increased toxicity rates were observed, necessitating dose reductions (5). Enzalutamide can be associated with cognitive impairment (6), and cabazitaxel has shown higher rates of adverse events in men older than 75 y (7).
In clinical studies, often only the fittest of elderly patients are included (8). Consequently, despite the fact that approximately 42% of cancer patients are at least 70 y old, they comprise less than 10% of the patients in clinical trials (9). [177Lu]-labeled prostate-specific membrane antigen (PSMA) radioligand therapy (RLT) has demonstrated a substantial antitumor effect and low rates of toxicities in mCRPC. However, in the pivotal VISION trial, only 1.7% (n = 14/831) of patients were 85 y or older (10).
The primary aim of this retrospective analysis was to assess the efficacy and safety of PSMA RLT in octogenarian mCRPC patients. Response and toxicity rates in chemotherapy-pretreated and chemotherapy-naïve patients were compared, and predictors for clinical progression-free survival (cPFS) and overall survival (OS) were investigated.
MATERIALS AND METHODS
Patients and Data Collection
mCRPC patients aged 80 y or older and treated with [177Lu]-PSMA-I&T RLT between October 2014 and February 2022 were identified from our institutional database. [177Lu]-PSMA-I&T was prepared according to good manufacturing practices and the German Medicinal Products Act (Arzneimittelgesetz §13 2b). All patients gave written informed consent and were treated under the conditions of Declaration of Helsinki article 37, “Unproven Interventions in Clinical Practice.” The responsible institutional review board approved the retrospective analysis (reference 115/18S).
Institutional eligibility criteria were previously published and were in accordance with the interdisciplinary German guideline for the treatment of prostate cancer after interdisciplinary tumor board decisions (11–13). Briefly, the study included mCRPC patients with pretreatment consisting of at least 1 line of androgen receptor–targeted agents and at least 1 line of taxane-based chemotherapy using a docetaxel or cabazitaxel regimen. Furthermore, patients who were ineligible for taxane-based chemotherapy because of poor Eastern Oncology Cooperative Group (ECOG) performance status (PS), higher-grade comorbidities, or refusal of chemotherapy were included in this retrospective data collection. Pretherapeutic PSMA PET was performed, and uptake was required to be higher in the tumor lesions than in the liver.
[177Lu]-PSMA RLT Procedure
Radiolabeling of [177Lu]-PSMA-I&T and administration of RLT were performed as previously reported (11,12,14). Patients usually underwent RLT at 6-wk intervals at a standard dose of 7.4 GBq. Treatment could be intensified with doses of up to 9 GBq or a 4-wk interval, or dosing could be reduced to 6 GBq depending on clinical factors such as organ function (e.g., impaired kidney or bone marrow function) and high or low tumor load, similar to other reports (15,16). The patients underwent PSMA PET/CT every 12 wk during treatment.
Intravenous treatment with [177Lu]-PSMA-I&T was continued up to a maximum of 12 cycles in patients with absence of radiographic or clinical progression and lack of severe toxicity. Androgen-deprivation therapy was continued during [177Lu]-PSMA-I&T RLT (Fig. 1). No other systemic treatments were allowed.
Swimmer plot of mCRPC treatments. Length of each bar symbolizes duration for which patient was on specific treatment.
Efficacy and Safety Assessment
Assessment of treatment efficacy was based on best prostate-specific antigen (PSA) response (PSA decline of ≥ 0%, ≥50%, or ≥90% during treatment). PSA progression-free survival (PFS) according to the recommendations of the Prostate Cancer Trials Clinical Working Group 3 (17) and OS were calculated. cPFS was defined as the time from treatment initiation to clinical or imaging progression or death, as previously published (12). Clinical progression included disease-related symptoms, reduced ECOG PS, and new cancer-related complications. Treatment-related toxicity was documented until 6 mo after the last RLT cycle and graded according to the Common Terminology Criteria for Adverse Events (version 5.0). Changes in ECOG PS after RLT were recorded. Assessment of renal function was based on the estimated glomerular filtration rate using the Chronic Kidney Disease Epidemiology Collaboration 2012 equation.
Statistical Analyses
Descriptive analyses were used to assess PSA response. The Kaplan–Meier curve method with corresponding 95% CI was used to analyze PSA PFS, cPFS, and OS. In a subgroup analysis, chemotherapy-naïve and chemotherapy-pretreated patients were compared for treatment outcome parameters.
Previously reported baseline parameters were used to evaluate potential predictions of cPFS and OS using the Cox proportional-hazards regression model. The following baseline variables were investigated: PS, PSA level, hemoglobin level, alkaline phosphatase level, lactate dehydrogenase (LDH) level, time between initial diagnosis and RLT initiation, presence of bone metastases, presence of visceral metastases, prior taxane-based chemotherapy, and primary metastatic disease. Only variables with significant differences from the univariate model (P < 0.05) were included in the multivariable analysis. Results are reported as hazard ratios (HRs) with 95% CIs.
A matched-pair subanalysis of 30 chemotherapy-naïve versus 30 chemotherapy-pretreated patients was conducted to adjust for possible bias at baseline. One patient of the chemotherapy-pretreated cohort could not be matched. Matching parameters were clinical and imaging-based parameters, consisting of time between first diagnosis and initiation of [177Lu]-PSMA-I&T RLT, initial Gleason score/International Society of Urological Pathology grade, primary metastatic disease, pretreatments for mCRPC, PSA at diagnosis and at RLT initiation, baseline AP and baseline LDH levels, and tumor distribution on baseline PSMA PET/CT. The patient characteristics of chemotherapy-naïve and chemotherapy-pretreated subgroups were compared using the Fisher exact test and the Mann–Whitney U test. Results were reported as frequency (percentage), median with 25th–75th percentiles, or median with range. A 2-sided P value of less than 0.05 was considered statistically significant. Statistical analysis was performed using SPSS software (version 24; IBM).
RESULTS
Patient Population and Treatment with [177Lu]-PSMA-I&T
Overall, 80 patients with mCRPC were analyzed. Their median age was 82 y (range, 80–91 y) at the initiation of [177Lu]-PSMA-I&T RLT. The median time between initial diagnosis and the start of RLT was 9 y (range, 1–25 y). Of 80 patients, 49 (61.3%) were chemotherapy-naïve. A median of 2 (range, 1–6) previous mCRPC treatment regimens had been applied, and 24 (30%) patients had received 3 or more mCRPC treatments before RLT. A swimmer plot was constructed to visualize mCRPC therapy lines (Fig. 1). Baseline characteristics, including site of metastases, baseline PSA levels, and details for the chemotherapy-naïve and chemotherapy-pretreated subgroups are presented in Table 1.
Baseline Patient Characteristics
In total, 324 cycles of [177Lu]-PSMA-I&T had been administered, with a median of 4 (interquartile range [IQR], 2–6) cycles per patient, a median dose of 7.3 GBq per cycle (IQR, 7.1–7.5 GBq), and a median cumulative activity of 23.8 GBq (IQR, 14.8–42.2 GBq). The median duration of RLT was 4 mo (IQR, 1–7 mo), and the median follow-up time was 11.4 mo (IQR, 5.8–16.7 mo). Supplemental Figure 1 shows details of the RLT cycles of the cohort (supplemental materials are available at http://jnm.snmjournals.org).
On the basis of the predefined criteria, 30 chemotherapy-pretreated patients could be matched to 30 chemotherapy-naïve patients from the total cohort of 80 patients. The baseline and treatment characteristics of the matched-pair subcohorts are shown in Supplemental Table 1.
Treatment Efficacy and Clinical Outcomes
A maximum PSA response of at least 30%, at least 50%, and at least 90% was achieved in 44 (55%), 37 (46.3%), and 22 (27.5%) of the patients, respectively (Fig. 2). A trend toward a higher PSA response in chemotherapy-naïve than in chemotherapy-pretreated patients was observed but was not statistically significant (e.g., ≥50% PSA decline in 51.0% vs. 38.7%, respectively; Supplemental Fig. 2). This tendency was confirmed in the matched-pair subanalysis (Supplemental Table 2).
Waterfall plot of best PSA response during [177Lu]-PSMA-I&T RLT. Blue and orange bars indicate chemotherapy-naïve and chemotherapy-pretreated patients, respectively. *PSA increase of >100%. CTx = chemotherapy; pts. = patients.
Median PSA PFS, cPFS, and OS were 6.1 mo (95% CI, 3.7–9.2 mo), 8.7 mo (95% CI, 5.9–10.5 mo), and 16.1 mo (95% CI, 13.3–20.9 mo), respectively. Median cPFS and median OS were significantly longer in chemotherapy-naïve patients than in chemotherapy-pretreated patients (10.5 vs. 6.5 mo and 20.7 vs. 11.8 mo; P = 0.01 and <0.01, respectively). A strong trend toward a longer PSA PFS in chemotherapy-naïve patients than in chemotherapy-pretreated patients was also observed but was not statistically significant (6.2 vs. 3.8 mo, respectively; P = 0.09) (Fig. 3). The results for PSA PFS, cPFS, and OS were comparable in the matched-pair subanalysis (Supplemental Fig. 2).
PSA PFS (A), cPFS (B), and OS (C) in chemotherapy-naïve patients (n = 49) vs. chemotherapy-pretreated patients (n = 31). CTx = chemotherapy.
Overall, 5 (6.3%) patients improved from ECOG PS 1 to 0. PS was stable in 56 (70.0%) patients (in 11 [13.8%] patients with ECOG 0, 37 [46.3%] patients with ECOG 1, and 8 [10.0%] patients with ECOG 2) after RLT. PS worsened by 1 and 2 grades in 14 (17.5%) and 4 (5%) patients after RLT, respectively. In 2 patients, RLT was stopped at the patient’s request, and in 7 patients RLT was stopped because of a worsening of the patient’s general condition or other reasons.
Predicting Factors in OS and cPFS
The results of the univariate and multivariable analyses are shown in Table 2. For OS, a shorter time from initial diagnosis (HR, 2.1 [95% CI, 1.0–4.2]; P = 0.04), a hemoglobin level below the median of 11.9 g/dL (HR, 2.8 [95% CI, 1.4–5.6]; P < 0.01), a LDH level above the median of 253 U/L (HR, 3.3 [95% CI, 1.6–6.9]; P < 0.01), and pretreatment with taxane-based chemotherapy (HR, 3.0 [95% CI, 1.6–5.7]; P < 0.01) were associated with a worse outcome on univariate analysis. However, in a multivariable Cox regression model, only a lower hemoglobin level (HR, 2.2 [95% CI, 1.1–4.4]; P = 0.02), a higher LDH level (HR, 2.3 [95% CI, 1.1–5.0]; P = 0.03), and prior taxane-based chemotherapy (HR, 2.0 [95% CI, 1.0–4.0]; P = 0.04) remained independent predictors of a shorter OS.
Univariate and Multivariable Analysis of OS and cPFS in Entire Cohort (80 Patients)
For shorter cPFS, significant independent predictors were a lower baseline hemoglobin level (HR, 2.2 [95% CI, 1.3–3.9]; P < 0.01) and a higher baseline LDH level (HR, 1.8 [95% CI, 1.0–3.2]; P = 0.04). A shorter time from initial diagnosis and pretreatment with taxane-based chemotherapy could not be confirmed as a significant predictor of a shorter cPFS on multivariable analysis. Univariate and multivariable analyses of OS and cPFS in the chemotherapy-naïve and chemotherapy-pretreated patients are presented in Supplemental Tables 3 and 4.
Toxicity
Treatment with [177Lu]-PSMA-I&T was well tolerated, with a low number of treatment-emergent grade 3 adverse events, such as grade 3 anemia in 4 (5%) patients (Table 3). No grade 4 adverse events occurred. The most frequent treatment-emergent low-grade adverse events were hematotoxicity, in particular anemia in 46% of the patients and thrombocytopenia in 16%. Renal toxicity of grade 3 or less, based on the estimated glomerular filtration rate changes, occurred in 20 (25%) of 80 patients, and grade 3 renal toxicity occurred in 4 (5%) patients. In 3 (3.8%) of 80 patients, a treatment-emergent decline in renal function led to termination of [177Lu]-PSMA-I&T RLT. The most frequent nonhematologic side effects were xerostomia of grade 2 or less and fatigue in 32% of the patients. The frequency of adverse events was similar between chemotherapy-naïve and chemotherapy-pretreated patients.
[177Lu]-PSMA-I&T RLT Emergent Toxicities
DISCUSSION
We retrospectively report on the treatment outcome and safety of [177Lu]-PSMA-I&T RLT in 80 mCRPC octogenarians. The patients underwent a median of 4 cycles with a favorable safety profile and a low frequency (<10%) of higher-grade toxicities. Rates of 50% PSA decline (46.3%), median PSA PFS (6.1 mo), and OS (16.1 mo) were comparable to those in a non–age-selected population. A more favorable outcome was observed in chemotherapy-naïve patients.
So far, studies on PSMA-targeted RLT report data on patients with a median age ranging between 70 and 72 y (10,12,16). Specifically, in the VISION trial only 1.7% (n = 14/831) of patients were at the age of 85 y or older (10). Because of reduced organ function and capability to tolerate side effects, oncologic treatments in senescent patients are challenging. For our retrospective analysis, we specifically selected octogenarians with a median age of 82 y at treatment initiation. In this selected group, treatment with [177Lu]-PSMA-I&T was overall safe and well tolerated, comparable to previously published data on RLT with [177Lu]-PSMA (10,12,16).
We acknowledge a high variation in reported toxicity rates in the literature for PSMA RLT in non–age-selected cohorts. Rates of anemia, for example, have varied: 19% of the patients in the TheraP-trial, 32% of the patients in the VISION trial, and 36% of the patients in a retrospective single-center study (10,12,18). The occurrence of posttherapeutic fatigue ranges between 20% and 70%, similar to xerostomia, which was observed in 24%–87% (10,12,16). These distinct variations most likely underline differences between retrospective and prospective studies and the limited comparability and standardization of subjective, patient-reported symptoms.
Given their reduced organ function, comorbidities, and impaired general condition, octogenarians are frequently designated as unfit for chemotherapy. We specifically investigated and compared side effects in chemotherapy-naïve and chemotherapy-pretreated patients and observed no substantial differences. A recent retrospective analysis of 83 chemotherapy-pretreated versus 84 chemotherapy-naïve non–age-selected mCRPC patients reported equal rates of side effects in both groups (19).
In comparison to taxane-based chemotherapy, the rates of side effects of [177Lu]-PSMA-I&T RLT are similar, with a trend toward a lower number of high-grade toxicities and lower rates of treatment discontinuation. A multicenter study of docetaxel as the first mCRPC treatment line in a comparable patient cohort of 123 octogenarians (median age, 82 y) reported toxicity with grade 3–4 neutropenia in 10.6%, fatigue in 9.7%, and diarrhea in 4.1%, but toxicity-related treatment discontinuation occurred in 15.4% (20). In a smaller cohort, Wong et al. reported distinctly higher rates of grade 3–4 hematologic toxicity in 9 of 20 (45%) patients and hospitalization of 5 patients because of chemotherapy-related complications (21). In the TROPIC trial, mCRPC patients older than 75 y and pretreated with docetaxel (n = 70) showed higher rates of diarrhea and neutropenia after cabazitaxel than did younger patients (n = 301) (55.7% and 24.2% vs. 44.5% and 17.6%, respectively) (7). Consistent with that, multivariable analysis of data from the European early-access program for cabazitaxel showed higher age to be a predictor of neutropenia of grade 3 or higher (22).
Similar observations with higher rates of toxicities in elderly patients have been reported for other mCRPC treatments. For the androgen receptor–targeting agent abiraterone, slightly higher rates of cardiac disorders such as atrial fibrillation and tachycardia were observed in mCRPC patients older than 75 y (5), and more frequent toxicity-caused treatment discontinuation was observed in octogenarians (23). For enzalutamide, similar tendencies were reported in elderly patients (24).
Impaired renal function is a known comorbidity in elderly patients (25). Currently, the impact of [177Lu]-PSMA RLT on renal function is not entirely clear. In early retrospective studies, no grade 3–4 renal toxicity was described (11). Assuming a similar delayed onset of kidney toxicity for [177Lu]-PSMA RLT, as known from, for example, peptide receptor radionuclide therapy, the current study protocols and the limited life span of mCRPC patients might not adequately capture its potential long-term effects. For instance, adverse events in patients in the pivotal phase III (VISION) trial were detected only up to 30 d after the last administration (10). However, our group recently reported 3 mCRPC patients showing histopathologically confirmed radiation nephropathy after high cumulative doses of [177Lu]-PSMA-I&T RLT (26). In the current study, 3 of 80 elderly patients eventually had to stop [177Lu]-PSMA-I&T RLT because of progressive renal impairment.
A PSA decline of at least 50% in 37 (46.3%) patients, a median PSA PFS of 6.1 mo, and a median OS of 16.1 mo are in the range of reports for a general mCRPC population and document the high antitumor effect of [177Lu]-PSMA RLT in elderly patients. Not unexpectedly, chemotherapy-naïve patients showed a significantly longer median cPFS and median OS of 10.5 and 20.7 mo, respectively. This finding was confirmed in a matched-pair subgroup analysis of 30 chemotherapy-naïve patients versus 30 chemotherapy-pretreated patients (median OS, 20.7 vs. 11.8 mo, respectively). Previously, Barber et al. reported similar differences for both groups (27.1 vs. 10.7 mo, respectively) in a general mCRPC cohort (19). These data compare well with docetaxel in a first-line mCRPC setting. In a similar patient cohort of 123 octogenarians (median age, 82 y), a PFS of 7 mo and a median OS of 20 mo were reported (20).
In our study, a lower hemoglobin level and a higher LDH level at treatment initiation were independent baseline predictors of both inferior cPFS and inferior OS, a finding that is in line with data on non–age-selected investigations (12,27). Likewise, prior taxane-based chemotherapy was associated with a poorer OS in a multivariable Cox regression model (19). The presence of visceral metastases, a significant predictor of OS in previous studies, was not related to OS in this analysis, possibly because of the limited number of patients.
The presented data have several limitations. The main limitation is the retrospective nature of the study, potentially missing side effects occurring outside the usual treatment and follow-up visits. Second, comparing rates of adverse events with data from the literature is prone to potential bias from different patient characteristics. Third, we could not perform further analysis with respect to the predictive nature of pretherapeutic PSMA PET, as patients in this cohort underwent PET with different PSMA ligands.
CONCLUSION
[177Lu]-PSMA-I&T RLT in mCRPC patients at least 80 y old is safe and effective, comparable to previously published data on non–age-selected cohorts with only a low rate of high-grade toxicities. Therefore, this treatment can be considered an alternative option in patients who are not eligible for taxane-based chemotherapy because of comorbidities or other relevant medical conditions. Similar to other agents, [177Lu]-PSMA-I&T RLT showed higher rates of biochemical response and a longer duration of response in chemotherapy-naïve patients. A prospective evaluation in earlier stages of mCRPC is already being explored in different studies (e.g., PSMAfore and SPLASH) to better define the role of [177Lu]-PSMA RLT in the therapy sequence of prostate cancer. The results might provide further insight into potential long-term toxicities (e.g., renal impairment), which are currently not fully known because of the limited life expectancy of mCRPC patients.
DISCLOSURE
Robert Tauber reports prior consulting activities for AstraZeneca, Bayer, BMS, Eisei, EUSA, Ipsen, Janssen, MSD, Philogen, Roche, and Sanofi; travel support from Bayer, BMS, Ipsen, Janssen, and Roche; and owning shares of Bayer. Hans-Jürgen Wester and Matthias Eiber are named in a patent application for rhPSMA and received funding from Blue Earth Diagnostics Ltd. (BED), Oxford, U.K. (licensee for rhPSMA), as part of an academic collaboration. Hans-Jürgen Wester is a founder, shareholder, and advisory board member of Scintomics GmbH, Fuerstenfeldbruck, Germany. Matthias Eiber reports prior consulting activities for BED, Novartis, Telix, Progenics, Bayer, Point Biopharma, and Janssen. Thomas Langbein reports consulting activities for BED. No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: The purpose of this retrospective analysis was to evaluate the efficacy and safety of [177Lu]-PSMA RLT in metastatic castration-resistant prostate cancer (mCRPC) patients at least 80 y old.
PERTINENT FINDINGS: In a retrospective analysis of patients at least 80 y old with mCRPC, [177Lu]-PSMA RLT showed comparable efficacy and safety to the published data. In the total cohort, 46.3% of the patients had a PSA response of at least 50%. In the subgroup analysis, there was a certain but nonsignificant advantage for patients without prior chemotherapy (PSA response, 51.0% vs. 38.7%). The therapy was well tolerated by the elderly patients. Critical grade 3 side effects occurred only rarely (e.g., anemia, 5%). No grade 4 toxicity was observed.
IMPLICATIONS FOR PATIENT CARE: [177Lu]-PSMA RLT seems to be a meaningful treatment option for older patients.
Footnotes
Published online Jun. 15, 2023.
- © 2023 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication December 6, 2022.
- Revision received March 14, 2023.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.