Long-term outcome of 177Lu-PSMA-617 radioligand therapy in heavily pre-treated metastatic castration-resistant prostate cancer patients

Objective Investigators have extensively explored the short-term safety and efficacy data on 177Lu-PSMA-617 radioligand therapy (RLT) in mCRPC patients. However, scarce literature is reported on the long-term outcome of these patients. The current goal of this study is focused on the long-term outcome of mCRPC patients treated with 177Lu-PSMA-617 RLT. Methods Among 135 patients, 121 mCRPC patients fulfilled the eligibility criteria and were included in the final analysis. Patients received a median of 3 cycles of 177Lu-PSMA-617 RLT at 6 to 12-week intervals. Primary endpoint included overall survival (OS) and secondary endpoints involved progression-free survival (PFS), predictive factors of OS and PFS, PSA response rate, molecular response, clinical response, and toxicity assessment. Results The median administered cumulative activity was 20 GBq (3.7–37 GBq). The median follow-up duration was 36 months (6–72 months). The estimated median PFS and OS were 12 months (mo) (95% CI: 10.3–13 mo) and 16 mo (95% CI: 13–17 mo), respectively. Any PSA decline and PSA decline >50% was achieved in 73% and 61% of the patients, respectively. Multivariate analysis revealed only failure to achieve >50% PSA decline as a significant factor associated with a poor PFS. Prognostic factors associated with reduced OS included, failure to experience >50% PSA decline, heavily pre-treated patient cohort who received >2 lines of prior treatment options, and patient sub-group treated with ≥2 lines of chemotherapy. Patients re-treated with additional treatment options after attaining 177Lu-PSMA refractory disease showed a remarkably prolonged OS. A significant clinical benefit was achieved post 177Lu-PSMA-617 RLT. The most common toxicities observed were fatigue (34.7%), followed by nausea (33%), and dry mouth (24.7%). Conclusion The current study supports the short-term safety and efficacy results of high response rates, prolonged PFS and OS, improved quality of life, and low treatment-related toxicities in patients treated with 177Lu-PSMA-617 radioligand therapy.


Patients
The study protocol was approved by the All India Institute of Medical Sciences, institute ethics committee  and conducted by the ethical standards declared by the Helsinki Convention. This study was funded by the Indian Council of Medical Research and registered on the Clinical Trial Registry-India (CTRI Ref No: CTRI/06/006998).
Patients were extensively discussed within the multidisciplinary tumor board following which they were referred for RLT to the Department of Nuclear Medicine. Patients were screened for the following eligibility criteria before the recruitment and written informed consent was signed by the patients.

Eligibility criteria for enrollment in the study
Essential criteria included pathologically confirmed prostatic adenocarcinoma, documented mCRPC status, progressive disease on standard treatment options including taxane-based chemotherapy, first-line and or second-line anti-androgen treatment, and/or androgen inhibitor therapies, patients on concomitant systemic treatments, documented disease progression on 68 Ga-PSMA PET/CT scan obtained within 28 days before the beginning of 177 Lu-PSMA-617 RLT with PSMA expression in lesions greater than the liver for soft tissue lesions and uptake greater than vertebra for the skeletal metastases and ECOG performance status up to 4. Patients were excluded if they had haemoglobin < 9 g/dL, absolute neutrophil count (ANC) <1. 5 x 109 /L, Platelets <60 x 109 /L, bilirubin >1.5 X upper normal limit (UNL), GFR < 40mL/min/1.73m 2 BSA), patients with serious comorbidities and medical conditions such as congestive heart failure, sensitivity to drugs, cord compression and those who denied giving written informed consent were excluded from the study.

Final patient recruitment
Among 135 consecutive mCRPC patients screened for eligibility, 14 patients dropped out for the following reasons: four had neuroendocrine differentiation of prostate cancer, six demonstrated low PSMA avidity on baseline 68 Ga-PSMA-11 PET/CT scan, three patients had grade III haematological toxicity associated with previous therapies, and one patient did not follow-up and dropped out after the first cycle of 177 Lu-PSMA-617 therapy on self-consent. Finally, 121 patients with a median age of 67 years who fulfilled the mandate criteria were included in the analysis. the early findings of our dosimetry report [6] showed no role of amino acid infusion for kidney protection in 177 Lu-PSMA-617 RLT. Hence, in the remaining 95 patients, only 177 Lu-PSMA-617 was diluted in 30 mL normal saline (0.9%), and administered by slow intravenous infusion over 5-10 minutes.

Dosing of 177 Lu-PSMA-617
A range of 1.11 to 7.8 GBq (30-210 mCi) of 177 Lu-PSMA-617 was administered per cycle in the patients. Due to the lack of evidence on the toxicity from 177 Lu-PSMA-617 RLT, at the initial phase of recruitment in 26 patients, we adopted a dose-escalation protocol which included dosages starting from 1.11 GBq and was gradually escalated up to 5.5 GBq of 177 Lu-PSMA-617 RLT. Among the 26 patients, 4, 13, 7, and 2 patients received 1.11, 1.85, 3.70 and 5.55 GBq of 177 Lu-PSMA-617 RLT, respectively. Dosimetry reports from various studies revealed doses up to 9.25 GBq safe for administration without adverse toxicities.
Hence, in the subsequent phase of the study, dosing of 177 Lu-PSMA-617 RLT ranged between 3.70 to 7.8 GBq at each cycle. The amount of administered activity was based on the extent of metastasis on the 68 Ga-PSMA-11 PET/CT scan, laboratory parameters, and the ECOG performance status. After the administration of 177 Lu-PSMA-617, all patients were admitted to the isolation therapy ward for clinical observation and eventually discharged when stable [5].

Follow-up
Patients were followed-up every 2, 4, and 6-8 weeks (wks) after each cycle of 177 Lu-PSMA-617 therapy with blood tests to assess for any treatment-related toxicities. After completion of the therapy regimen or discontinuation of treatment, patients were followed regularly at monthly intervals with PSA values and 68 Ga-PSMA PET/CT at both Nuclear Medicine and Medical Oncology clinics.

Treatment outcome measures
The primary outcome measure included overall survival. Secondary outcome measures involved progression-free survival, factors predicting the OS and PFS, evaluation of the PSA response rate, molecular response assessment, clinical response assessment, and adverse event profile.
Survival analysis. Overall survival. Time from the commencement of 177 Lu-PSMA-617 therapy to the death due to any cause or the date of the last contact.
Progression-free survival. Defined as the time from the initiation of treatment to the date of documented disease progression (PSA rise >25% from the baseline value, which was further confirmed with a repeat 3-week value). Patients without any tumor progression at the time of analysis were censored at their last tumor evaluation date.
Biochemical response. Biochemical response to treatment was assessed as per the Prostate Cancer Working Group 3 criteria (PCWG3) [7].
PSA response rate. Defined as the percentage of patients with a PSA reduction of >50% from baseline. Assessed at 2, 4, and 8 wks after every cycle of 177 Lu-PSMA-617 RLT with repetition at monthly intervals after completion of the 177 Lu-PSMA-617 treatment regimen.
Molecular tumor response. The molecular response was assessed using PERCIST 1 criteria [8]. 68 Ga- PSMA-11 PET/CT scan for the morphological and molecular response was repeated  after three cycles of 177 Lu-PSMA-617 treatment or only if deemed necessary in situations such  as subsequent post-therapy scan showed either exceptional response or appearance of the new  lesion, new sites of pain development, decrease or doubling of PSA and during the follow-ups  after discontinuation or completing the 177 Lu-PSMA RLT regimen. Clinical response. Clinical response criteria included the assessment of visual analog score (VAS) [9], analgesic score (AS) [9], Karnofsky performance status (KPS) [10], and Eastern Cooperative Oncology Group (ECOG) performance status.
Toxicity. Treatment-related adverse events (AEs) were documented as per the National Cancer Institute for Common Toxicity Criteria version 5.0 [11].

Discontinuation of 177 Lu-PSMA-617 therapy
Treatment was ceased in the following circumstances: Progression of disease, an early response to treatment which was attained before completion of the treatment regimen, no improvement in the quality of life with no clinical benefit after board discussion with the treating medical oncologist, achieved the maximum administered activity dose limit of~37 GBq (1000 mCi)

Statistical analysis
Stata v11.2 statistical software (StataCorp, College Station, TX, USA) and Medcalc ver-sion12.5.0 were used to perform the analysis. The D'Agostino-Pearson test was conducted to assess the normality of the data. The normal distributed continuous data were presented as mean, standard deviation, and range. Skewed data were depicted as the median and interquartile range (IQR). Paired-samples t-test (parametric) or Wilcoxon signed-rank test (nonparametric) tests were used to compare the pre and post-therapy parameters.
Kaplan-Meier survival curves were generated, and the Log-rank test was used to compare the OS and PFS of categorized variables. Their median values designated the cut-offs for continuous variables. All variables with P value �0.1 on univariate analysis Cox proportional-hazards regression model were included in the multivariate model. For multivariate analysis, the Cox proportional-hazards regression model by stepwise elimination method was carried out to determine the prognostic factors associated with OS and PFS. P-values <0.05 were considered significant.

Results
Out of 135 mCRPC patients, 121 patients fulfilled the mandate criteria and were included in the analysis. Among the entire series, the first 177 Lu-PSMA-617 therapy cycle was administered in September 2014, and the last patient was recruited in February 2020. The cut-off date for follow-up was September 2020. According to the reverse Kaplan-Meier censoring method, the median follow-up duration was 36 mo (6-72 mo) from the start of 177 Lu-PSMA-617 RLT. The baseline characteristics of patients are displayed in Table 1. Fourteen patients had a family history of various cancers and three among them had a family history of prostate cancer. Fortyeight percent belonged to Gleason score 9-10 with 35% having extensive marrow/skeletal involvement and a median ECOG status of 3. Ninety-one percent of (110/121) patients were treated with more than two lines of standard therapies. The remaining 9% had at least two treatment lines before RLT; 83.4% of patients had received docetaxel, 53.7% (65/121) had androgen synthesis inhibitor therapy, and 19% (23/121) received enzalutamide. The median duration of first-line and second-line treatment has been detailed in Table 1.
PSMA expression was noted in the primary site in 73% of patients. Consistent with the advanced stage of CRPC, all patients in this series presented with skeletal metastases; however,

Dose scheme and treatment cycles
In 2014, we started the 177 Lu-PSMA-617 RLT when there were no dosimetry reports available for the dose-limiting organs. Thus, the initial 26 patients had a dose-escalation study for dosimetry purposes. In the remaining 95 patients, the 177 Lu-PSMA-617 administration was an individualized approach. It varied among the patients between 3.70 and 7.78 GBq per cycle. A median of 3 cycles of 177 Lu-PSMA-617 was administered with the mean cumulative activity of 20 GBq (range, 3.7-37 GBq). In a median time interval of 8 wks (range, 6-12 wks) between each 177 Lu-PSMA-617 RLT treatment cycle, a total of 386 cycles were administered in 121 mCRPC patients. The number of cycles varied from 2 cycles in 44 (36%), 3 cycles in 38 (32%), and 4 cycles in 16 (13%). Furthermore, 21 (17%) patients received more than four cycles. Only two patients received a single cycle of 177 Lu-PSMA-617 RLT due to the progression of diseases but were on regular follow-up. The survival status of each patient according to the number of cycles administered is depicted in the flow-chart (Fig 1).

Biochemical PFS, OS, and predictive factors
During the follow-up, 79 (65%) patients experienced disease progression with a median PFS of 12 mo (95% CI: 10.3-13), and a 24-months progression-free survival probability of 19.7% (Fig  2A). Thirty-eight patients in this series, who experienced >50% PSA decline during the treatment, eventually demonstrated disease progression during the follow-up period. The univariate analysis revealed no prior history of androgen-inhibitor treatment chemotherapy naïve, patients experiencing any PSA decline and/or >50% PSA decline as significant predictors of prolonged PFS (Table 2). However, multivariate, stepwise, Cox proportional hazard regression analysis revealed only failure to demonstrate more than 50% PSA decline (hazard ratio [HR],  (Fig 3A).
On a detailed analysis of the various regimes of treatment that were opted in patients after either on progression during 177 Lu-PSMA-617 therapy or during the follow-up period after completion of 177 Lu-PSMA-617 therapy cycles, 27 patients underwent further salvage treatment options that are detailed in Table 4. Interestingly, 13 of these 27 patients are still alive and are responding to their current treatments (Table 4). A sub-categorical analysis revealed a remarkably improved OS in these patients treated with alternative treatment options even after progression on 177 Lu-PSMA-617 therapy compared to those who did not receive further treatment. [14 vs. 22 mo, HR: 0.4, 95% CI: 0.224-0.549; P<0.0001] with consistent significance in the multivariate Cox-proportional hazard model (Fig 3E).

Biochemical PSA response assessment
During the treatment and follow-up, 88 (73%) patients experienced any PSA decline, and the best PSA response of >50% PSA decline was seen in 74 (61%) patients.
During the treatment or follow-up, >25% PSA progression was observed in, 79 (65%) patients. Among them, 38 patients had shown >50% PSA response during some point of the treatment and follow-up.

Molecular response
Interim and the follow-up molecular response assessment could be assessed in 105 patients. Three patients achieved complete response during the treatment and remained disease-free till their end-follow-up. Out of 27 patients who demonstrate initial PR, two achieved complete remission, eight patients further responded to treatment (NCR/PR), and 17 patients progressed. Thirty-three patients demonstrated molecular disease progression at their first interim 68 Ga-PSMA PET/CT scan and did not respond to further treatment (Table 5).

Clinical response
There was a significant reduction in the VASmax score post-treatment (8 vs. 5; P < 0.0001). Similarly, a significant decrease in mean AS from the baseline was observed (3 vs. 2 ± 1 [P<

Toxicity and adverse events
The Other toxicities such as diarrhea were reported in 14% 18/121) of patients. Grade III anemia was noted in three patients. Interestingly, there was no grade III thrombocytopenia or leukopenia in any of the patients. Regarding nephrotoxicity, two patients with grade I kidney toxicity at the baseline marginally increased to grade II toxicity, and no higher-grade toxicity was observed (Table 6).

Discussion
This study aimed to give an overview on the long-term outcome of 177 Lu-PSMA-617 therapy in a heavily pre-treated mCRPC patient cohort, who had received at least two lines of prior treatment; particularly, focused on the survival outcomes and the prognostic factors that influence the survival. In this study, we present a single institutional 6-year experience by highlighting the long-term outcome of 177 Lu-PSMA-617 therapy.
To the best of our knowledge, there is only one study by Violet et al. [4] and ours would be the second to elaborate the long-term outcome of patients treated with 177 Lu-PSMA-617 RLT with a median follow-up duration of 36 months. Similarly, both the studies narrate the response and pattern of disease progression in patients during treatment and post-treatment follow-up. However, while we studied the outcome of 177 Lu-PSMA-617 therapy where patients were further treated with various systemic salvage therapeutic options, Violet et al. [4] re-challenged the second set of

PLOS ONE
Long-term outcome of 177 Lu-PSMA-617 in mCRPC patients 177 Lu-PSMA-617 therapy regimen in their patient cohort. In this current study, we report the pattern of progression, the time to disease progression after the initial response, and the further course of treatment with additional systemic anti-cancer/salvage treatment options.
Violet et al. [4] reported a >50% PSA decline rate of 64% when re-treated with 177 Lu-PSMA-617 RLT, which is well in line with the PSA response rate observed in the current study in 61% of patients. The long-term PSA response rates concur well with both previously published retrospective and prospective short-term results [1]. However, when comparing our results to those of older studies, it must be pointed out that the PSA response rates were in the lower range. A recently published meta-analysis in 744 mCRPC patients on 177 Lu-PSMA-617 reported a pooled >50% PSA decline rate of 46% (95% CI, 40-53%) [1]. The primary outcome measure in this study was the overall survival. Violet et al. [4] observed a median OS of 13.3 mo (95% CI: 10.5-8.5) compared to 16 mo in our study. In agreement, meta-analysis data pooled from the previously published works addressed a median overall survival of 13.7 (IQR: 8-14) mo [1]. Moreover, the long-term OS data were similar to and consistent with the short-term survival data.
The secondary objective of our study was to identify the prognostic factors influencing overall survival. The association of PSA response with the overall survival has been previous studied with variable results. We observed that a PSA response of >50% is a significant predictor of prolonged OS. A similar result was reported by Violet et al. [4]. However, in short-term studies, the impact of >50% PSA decline on the OS remains inconsistent with few studies favoring it as a critical prognostic indicator of OS, whereas some other studies contradicting the fact [12,13,[15][16][17]. This inconsistency in data leads to an interesting finding that a PSA response at the initial phase does not always translate to a longer OS in mCRPC patients with aggressive metastases. In contrast, both long-term studies concurred that PSA response of >50% decline predicts prolonged OS. Our findings on OS contradict the results of Brauer et al. [13] and Ahmadzadehfar et al. [12]. However, a later study by Ahmadzadehfar et al. [14] reported a decline in PSA levels of >50% significantly attributed to the longer median OS (70 wks; 95% CI: 39.5-100.5) compared those patients with PSA decline <50% (49 wks; 95% CI: 30.2-67.8). The discrepancy in the two results may be due to the selection bias, small sample size, heterogeneity in the disease burden, and probably shorter duration of follow-up (Table 7).  One of the key findings from this study indicates that the overall survival remarkably improved in the patient cohort who were eventually treated with other anti-cancer treatment options after the discontinuation of 177 Lu-PSMA-617 RLT. These findings correlate well with Violet et al. [4] and further favours the concept of re-challenging patients with other advanced salvage treatment options after acquiring radio-resistance on 177 Lu-PSMA-617 RLT.

Influence of prior lines of treatment on OS
An important observation noted in the current study is that patients who were subjected to more than two prior systemic lines of treatment options showed poor overall survival. In this context, Ahmadzadehfar et al. [18] retrospectively evaluated the impact of prior therapies on OS in 416 patients treated with 177 Lu-PSMA-617 from 11 different clinics. The multicentre analysis revealed prior chemotherapy as a significant independent adverse prognostic factor for poor overall survival. Similar to the findings of Ahmadzadehfar et al. [18], both univariate and multivariate analyses demonstrated prior treatment with >2 lines of therapies, and a prior history of 2 lines of chemotherapy significantly reduced OS. Similarly, Kessel et al. [17] observed second-line cabazitaxel chemotherapy (6.7 vs. 15.7 mo, P = 0.002) significantly associated with a shorter OS compared to patients who had not received second-line chemotherapy (7.9 vs.14.6 mo, Log-rank P = 0.002; HR 2.1, P = 0.009). From the above data, it is clear that heavily pre-treated patients with multiple lines of treatment negatively impact the overall survival ( Table 7).

Influence of concomitant treatment on OS
Several clinical trials on concomitant treatment with 177 Lu-PMSA-617 are in progress. Although a prior anti-hormonal therapy with either abiraterone or enzalutamide or both was not a significant predictive factor, there was a significant difference between the OS of patients with a history of enzalutamide and patients who were under concurrent usage of enzalutamide during 177 Lu-PSMA-617 treatment (12.3 vs. 10.8 mo, respectively; P = 0.045) Combination therapy, especially regarding second-line androgen deprivation therapy or chemotherapy, was assessed in our patient population, and its use did not affect the OS in the long-run. The short-term results of our previous study also demonstrated concomitant medication prolonged the survival but could not remain a predictive factor on multivariate analysis. However, these findings are preliminary, and the type of concomitant medications varied across the patients. But definitive clinical trials are in progress to evaluate the impact of concomitant treatment on 177 Lu-PSMA-617 therapy.

Re-treatment cohort
Twenty-seven patients with disease progression further underwent other lines of treatment, among which 16 patients primarily received 225 Ac-PSMA-617 therapy and have shown remarkable response rate. Fifty-six percent of the patients are alive and have shown improved overall survival. Recently, a study on 225 Ac-PSMA-617 in mCRPC patients from our group has shown a promising disease control-rate (DCR), and minimal treatment-related toxicities in patients refractory to 177 Lu-PSMA-617 therapy [19]. Eight patients also received enzalutamide among whom 3 are alive ( Table 4). The impact of further treatment post after 177 Lu-PSMA-617 may be the underlying fact for improved survival of 22 mo compared to 14 mo in the retreatment naïve cohort.

Treatment-related toxicities
The long-term follow-up of these patients after the completion of 177 Lu-PSMA-617 RLT will best answer the delayed complications. The results of our study reassure the earlier findings of short-term observations regarding the safety of RLT; the toxicities were minimal and rare chances of delayed toxicity to the critical organs such as the kidneys and salivary glands. There is no adverse risk noted over several years of meticulous follow-up of these patients to the dose-limiting organs. Similarly, a study on the toxicity of the re-challenged patient cohort with 177 Lu-PMSA-617 reported no unexpected adverse events with 177 Lu-PSMA-617 retreatment [4]. Keeping this in mind, higher fixed doses of administered activity per cycle and/or re-challenge of 177 Lu-PSMA-617 therapy regimen may be seriously considered. In the future, even administering concurrent therapies in these patients is feasible. 177 Lu-PSMA-617 therapy is safe with limited toxicities, even in heavily pre-treated mCRPC patients with extensive bonemarrow involvement. The pancytopenia was limited to grade 2, transient with a nadir around four weeks of therapy. Less serious, but more common, side effects included xerostomia, fatigue, and nausea.

Limitations
The study suffers from certain limitations. Firstly, it's a retrospective and single-center study. The study population was heterogeneous with a wide range of ECOG performance status, variation in the pattern of metastases and, heterogeneities in the prior therapies due to the broad inclusion criteria. Hence the treatment protocol was individualized and may suffer from heterogeneity in the treatment and outcomes. Genomic data was not assessed and hence, a clinico-genetic correlation was out of the scope of this paper.

Conclusion
From this long-term follow-up study, we conclude that 177 Lu-PSMA-617 RLT is an effective form of therapy in mCRPC patients and confirms the findings of the short-term studies. 177 Lu-PSMA-617 prolongs the overall survival in mCRPC patients heavily pre-treated with chemotherapy, first and second-line anti-androgens, and androgen inhibitor therapies. The delayed toxicity is low and acceptable by most of these patients. Patients who were treated with various other mCRPC-approved compounds before 177 Lu-PSMA-RLT were associated with a significantly shorter OS than that of the patients who received 177 Lu-PSMA RLT at the earlier stages of mCRPC. An aggressive treatment approach, either sequential or in combination with second-generation anti-androgens, chemotherapies, or with investigational 225 Ac-PSMA-617 alpha therapies, might prolong the survival of mCRPC patients in the future.