Review Reports - Implementation and Real-World Outcomes of [177Lu]Lu-PSMA-617 for Metastatic Androgen Pathway Modulation-Resistant Prostate Cancer: Following Public Reimbursement in Canada

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript presents a well-structured and clinically relevant retrospective study of [¹⁷⁷Lu]Lu-PSMA-617 therapy following public reimbursement in Canada. The results are consistent with previous clinical studies and provide valuable evidence regarding the implementation of PSMA-targeted radioligand therapy in routine clinical practice. The discussion is generally balanced and appropriately addresses the main limitations.

A few points would benefit from clarification:

Radiopharmaceutical preparation: Please clarify whether the commercially available Pluvicto (Novartis) or a locally manufactured [¹⁷⁷Lu]Lu-PSMA-617 preparation was administered.
Health-economic considerations: Given the emphasis on public reimbursement, the discussion should briefly address published cost-effectiveness analyses, particularly the Canadian CADTH assessment and the German analysis by Brinkmann et al. (2025). The incremental costs per quality-adjusted life-year gained are particularly relevant to the implementation of this therapy.
Treatment access: The median interval of 47 days between PSMA PET/CT and treatment initiation is noteworthy. Were treatment capacity limitations or delays responsible for excluding otherwise eligible patients?
Clinical outcomes: The relatively short median follow-up of 7.2 months should be considered when interpreting the estimated median overall survival of 12.3 months.
Future perspectives: A brief discussion of emerging PSMA-targeted radioligand therapies involving actinium-225 and terbium-161 would provide a useful perspective on future therapeutic developments.

Overall, this is a valuable contribution to the clinical literature, with findings that are relevant to both routine patient care and the broader implementation of PSMA-targeted radioligand therapy.

Author Response

A few points would benefit from clarification:

Radiopharmaceutical preparation: Please clarify whether the commercially available Pluvicto (Novartis) or a locally manufactured [¹⁷⁷Lu]Lu-PSMA-617 preparation was administered.

Response:
We thank the Reviewer for the opportunity to clarify treatment source was from commercial supply from Novartis. We added “doses were obtained from Novartis commercial supply” to Materials and Methods section (page 3).

Health-economic considerations: Given the emphasis on public reimbursement, the discussion should briefly address published cost-effectiveness analyses, particularly the Canadian CADTH assessment and the German analysis by Brinkmann et al. (2025). The incremental costs per quality-adjusted life-year gained are particularly relevant to the implementation of this therapy.

Response:
We thank the Reviewer for highlighting this important point. We added the following in the Discussion and cited the two suggested references.

“Although cost-effectiveness was beyond the scope of this study, prior analyses have highlighted the substantial cost associated with [¹⁷⁷Lu]Lu-PSMA-617, which exceeds the Canadian benchmark willingness-to-pay threshold of $50,000 per quality adjusted life-year (QALY) gained [21], [22]. Requirement for PSMA PET imaging may further increase health system costs. Canada’s Drug Agency (CADTH) recommended reimbursement with conditions, including reduced drug prices, based on the survival and quality of life benefits demonstrated in VISION. Real-world cost effectiveness following implementation warrants further evaluation.” (page 11)

Treatment access: The median interval of 47 days between PSMA PET/CT and treatment initiation is noteworthy. Were treatment capacity limitations or delays responsible for excluding otherwise eligible patients?

Response:
We thank the Reviewer for this important question. Treatment capacity at Nuclear Medicine was not a limitation and was not a reason for excluding eligible patients.

We added in the Discussion “Treatment capacity was not a limiting factor at our centre. However, some patients underwent PSMA PET/CT before radiographic or clinical progression necessitating a change in systemic therapy, contributing a longer interval before treatment initiation.” (page 9)

Clinical outcomes: The relatively short median follow-up of 7.2 months should be considered when interpreting the estimated median overall survival of 12.3 months.

Response:
We thank the Reviewer for highlighting this important point. We added in the Discussion “Given the relatively short follow up, OS data in our cohort may be immature” (page 10)

Future perspectives: A brief discussion of emerging PSMA-targeted radioligand therapies involving actinium-225 and terbium-161 would provide a useful perspective on future therapeutic developments.

Overall, this is a valuable contribution to the clinical literature, with findings that are relevant to both routine patient care and the broader implementation of PSMA-targeted radioligand therapy.

Response:
We thank the Reviewer for this excellent suggestion. We have added the following in the Discussion.

“Real-world cost effectiveness following implementation warrants further evaluation, particularly as the PSMA radioligand therapy landscape expands. As emerging PSMA-targeted radioligand therapies, including Actinium-225, Terbium-161, and Lead-212, advance in development, experience with the implementation of [¹⁷⁷Lu]Lu-PSMA-617 may help inform their integration into publicly funded healthcare systems [23].” (page 11)

Reviewer 2 Report

Comments and Suggestions for Authors

General Comments

The manuscript presents a retrospective, single-center study investigating the clinical outcomes of [¹⁷⁷Lu]Lu-PSMA-617 radioligand therapy in 77 patients with metastatic castration-resistant prostate cancer (mCRPC) treated within a publicly funded healthcare system in Ontario, Canada.

The topic is clinically relevant, providing real-world evidence on the implementation of PSMA-targeted radioligand therapy in view of its increasing use and the organizational challenges associated with implementing theranostic programs in routine clinical practice.

The study provides useful real-world data regarding treatment delivery, PSA and radiographic responses, outcomes, and treatment-related toxicity. The inclusion of a heterogeneous patient population, potentially less selected than those enrolled in prospective clinical trials, represents an important strength.

However, several important methodological aspects require clarification. In addition, inconsistencies in the numerical results and supplementary materials must be resolved to ensure the reliability and interpretability of the findings. Addressing these issues would substantially improve the scientific quality, methodological transparency, and clinical relevance of the manuscript.

Major Comments

The authors report that patients were selected for [¹⁷⁷Lu]Lu-PSMA-617 therapy based on PSMA PET/CT findings fulfilling the VISION eligibility criteria, while FDG PET/CT was not routinely performed. Although this approach is consistent with the selection strategy used in the VISION trial, the absence of FDG PET/CT deserves further discussion, particularly considering the potential presence of discordant FDG-positive/PSMA-negative lesions in patients with advanced mCRPC. The authors should clarify whether FDG PET/CT was omitted exclusively because it was not required by the reimbursement criteria or whether additional considerations influenced this decision. Furthermore, it would be useful to discuss whether patients with aggressive clinical features, visceral metastases, or suspected dedifferentiated disease were considered for additional FDG PET/CT assessment.
The manuscript states that baseline PSMA PET/CT was performed using either [⁶⁸Ga]Ga-PSMA-11 or [¹⁸F]DCFPyL. The authors should specify the number of patients examined with each radiotracer. Furthermore, the application of the VISION eligibility criteria should be described in greater detail, particularly regarding the identification of PSMA-negative lesions on conventional imaging.
The authors report that restaging was generally performed after 2–4 treatment cycles using contrast-enhanced CT and bone scintigraphy. Considering that PSMA PET/CT was used for baseline patient selection and provides molecular information on PSMA-expressing disease, it would be valuable to discuss why it was not routinely repeated during and after treatment. Although conventional imaging is consistent with established trial-based response assessment, its use should be contextualized within the theranostic setting.
Clarification of post-therapy SPECT/CT examinations. The authors state that "SPECT CT were obtained every 2 cycles whenever feasible", but the purpose and acquisition protocol of these examinations are not described. Please clarify whether these were post-therapy [¹⁷⁷Lu]Lu-PSMA-617 SPECT/CT acquisitions performed to evaluate radiopharmaceutical biodistribution. If so, the authors should specify the anatomical coverage, and the number of patients who underwent these examinations. Furthermore, it should be clarified whether post-therapy SPECT/CT findings were used exclusively for biodistribution assessment or whether they contributed to treatment response evaluation and clinical decision-making. A more detailed description of this aspect would strengthen the nuclear medicine and theranostic relevance of the manuscript.
The manuscript states that radiographic response was predominantly assessed using conventional imaging (94%), whereas only two patients underwent post-treatment PSMA PET/CT. The authors should clarify whether progression was defined on CT, bone scintigraphy, PSMA PET/CT, or a combination of these modalities, and whether standardized criteria were consistently applied across the cohort. In particular, the use of RECIST 1.1 and PCWG3 criteria should be described more precisely. Importantly, the authors should discuss whether the use of different imaging modalities and nonuniform imaging follow-up could have influenced the detection of disease progression and the estimation of radiographic progression-free survival.
The manuscript describes a single-center retrospective study conducted at the University Health Network (UHN), while several sections refer specifically to the Princess Margaret Cancer Centre (PM). The authors should clarify the relationship between these institutions and specify the actual clinical setting in which the study was conducted and the treatment was administered.
Clarification of the study population and Discrepancy: the manuscript reports different patient numbers across the Results section and supplementary materials, including 77, 60, and 59 patients. Although these values appear to refer to different populations, their definitions and relationships are not sufficiently clear. The authors should clearly explain the difference between these populations. In addition, the radiographic response population includes 64 patients in the main manuscript and 53 patients in Supplementary Table S2. A patient flow diagram illustrating the overall cohort, treatment completion status, and the number of patients evaluable for each outcome would considerably improve the clarity of the manuscript.
An important discrepancy was identified regarding the number of patients achieving a PSA50 response. The main manuscript reports that 31 of 77 patients (40.3%) achieved a PSA50 response, whereas Supplementary Table S4 reports 33 patients with PSA50 response. The authors should carefully review the original dataset and ensure that the number of patients achieving PSA50 response is reported consistently throughout the abstract, Results section, figures, tables, and supplementary materials.
Several additional discrepancies are present between the main manuscript and the supplementary tables. Some subgroup counts within Supplementary Tables should be rechecked. The authors are therefore encouraged to conduct a comprehensive verification of all reported data, including patient numbers, percentages, subgroup analyses, and statistical results.

Minor comments

The abbreviation "GU" in "GU medical oncologists" should be defined at first mention as "genitourinary (GU) medical oncologists".
The supplementary tables are referred to as "Appendix Tables A1–A10" in the main manuscript, whereas the supplementary document labels them as "Tables S1–S10". The numbering should be standardized throughout the manuscript to ensure consistency and facilitate cross-referencing.
The Supplementary Materials section ("The following supporting information can be downloaded at: https://www.mdpi.com/article/doi/s1, Table S1...; Table S2: ; Table S3: ...") appears incomplete.
Radiopharmaceutical nomenclature should be standardized throughout the manuscript. The notation [¹⁷⁷Lu]Lu-PSMA-617 is recommended for consistency.
The title and main text should consistently use the term "castration-resistant prostate cancer" rather than "castrate resistant prostate cancer".

Comments on the Quality of English Language

Minor grammatical and typographical errors should be corrected throughout the manuscript, and all tables, figures, and abbreviations should be checked for consistency.

Author Response

Major Comments

The authors report that patients were selected for [¹⁷⁷Lu]Lu-PSMA-617 therapy based on PSMA PET/CT findings fulfilling the VISION eligibility criteria, while FDG PET/CT was not routinely performed. Although this approach is consistent with the selection strategy used in the VISION trial, the absence of FDG PET/CT deserves further discussion, particularly considering the potential presence of discordant FDG-positive/PSMA-negative lesions in patients with advanced mCRPC.

The authors should clarify whether FDG PET/CT was omitted exclusively because it was not required by the reimbursement criteria or whether additional considerations influenced this decision.

Response:
We thank the Reviewer for raising these very important points. FDG PET/CT was not required by the reimbursement criteria, however it was not meant to be omitted exclusively.

We added the following to the Materials and Methods section “FDG PET/CT was not routinely required as per provincial reimbursement criteria, but could be obtained by the treating oncologist if felt clinically indicated.” (page 3)

Furthermore, it would be useful to discuss whether patients with aggressive clinical features, visceral metastases, or suspected dedifferentiated disease were considered for additional FDG PET/CT assessment.

Response:

We further added to the discussion on this topic: “Our institutional theranostic multidisciplinary cancer conference recommends its consideration in patients with borderline PSMA avidity, or features suggestive of aggressive or dedifferentiated disease such as low PSA production and predominant visceral metastases. Biopsy of discordant FDG-positive/PSMA-negative disease is recommended when feasible to exclude neuroendocrine differentiation. Routine incorporation of FDG PET/CT into eligibility assessment, however, must be balanced against the additional resource requirements and potential delays in treatment initiation within a publicly funded healthcare system.” (page 9-10)

The manuscript states that baseline PSMA PET/CT was performed using either [⁶⁸Ga]Ga-PSMA-11 or [¹⁸F]DCFPyL. The authors should specify the number of patients examined with each radiotracer. Furthermore, the application of the VISION eligibility criteria should be described in greater detail, particularly regarding the identification of PSMA-negative lesions on conventional imaging.

Response:
We thank the Reviewer for this inquiry. We added the following on page 5:

Among 77 patients, 14 patients had [¹⁸F]DCFPyL tracer, 63 patients had [⁶⁸Ga]Ga-PSMA-11.
No patients had sizable PSMA-negative lesions (osseous metastases with a ≥1 cm soft-tissue component, visceral metastases ≥1 cm, or nodal metastases ≥2.5 cm), defined as PSMA uptake below hepatic background on corresponding conventional imaging.

The authors report that restaging was generally performed after 2–4 treatment cycles using contrast-enhanced CT and bone scintigraphy. Considering that PSMA PET/CT was used for baseline patient selection and provides molecular information on PSMA-expressing disease, it would be valuable to discuss why it was not routinely repeated during and after treatment. Although conventional imaging is consistent with established trial-based response assessment, its use should be contextualized within the theranostic setting.

Response:
We thank the Reviewer for the opportunity to clarify. Repeat PSMA PET/CT for response assessment was not routinely available; therefore, treatment response was generally evaluated using conventional imaging. We agree the use of serial PSMA PET/CT should be contextualized as universal standards for response criteria has not been established to our knowledge.

We added the following

in the Material and Methods section after outlining restaging with conventional imaging after 2-4 cycles: “Repeat PSMA PET for response assessment to [¹⁷⁷Lu]Lu-PSMA-617 was not routinely available.” (page 3)
in the Discussion section “While most patients did not undergo serial PSMA PET/CT, further studies are needed to define its role in response assessment.” (page 10)

Clarification of post-therapy SPECT/CT examinations. The authors state that "SPECT CT were obtained every 2 cycles whenever feasible", but the purpose and acquisition protocol of these examinations are not described. Please clarify whether these were post-therapy [¹⁷⁷Lu]Lu-PSMA-617 SPECT/CT acquisitions performed to evaluate radiopharmaceutical biodistribution. If so, the authors should specify the anatomical coverage, and the number of patients who underwent these examinations. Furthermore, it should be clarified whether post-therapy SPECT/CT findings were used exclusively for biodistribution assessment or whether they contributed to treatment response evaluation and clinical decision-making. A more detailed description of this aspect would strengthen the nuclear medicine and theranostic relevance of the manuscript.

Response:
We thank the Reviewer for raising this important point. SPECT CTs were post therapy acquisitions. However post therapy SPECT CT data were not systematically collected or analyzed as part of the present study and therefore detailed parameters and requested information are unfortunately not available at this time. These examinations were primarily used to assess post-treatment radiopharmaceutical distribution and were not routinely used for formal response assessment or to guide treatment decisions. Post-therapy SPECT/CT data collection is currently being undertaken separately with our Nuclear Medicine team.

To align with Reviewer 3’s suggestions, we have decided to remove the SPECT CT section as we don’t have results to present at this time (page 3)

The manuscript states that radiographic response was predominantly assessed using conventional imaging (94%), whereas only two patients underwent post-treatment PSMA PET/CT.

The authors should clarify whether progression was defined on CT, bone scintigraphy, PSMA PET/CT, or a combination of these modalities, and whether standardized criteria were consistently applied across the cohort.

Response:
We appreciate these suggestions by the Reviewer. Radiographic response was assessed using conventional imaging to align with the VISION trial and other real-world cohorts. We added in the Material and Methods section “conventional imaging” (page 3) and in the Discussion “defined by conventional imaging” (page 10)

In particular, the use of RECIST 1.1 and PCWG3 criteria should be described more precisely.

Response:
Thank you for the opportunity to clarify. We added the following to the Methodology section (page 3-4). “In summary for RECIST 1.1, up to five target lesions per metastatic site were recorded as target lesions. Complete response (CR) was defined as disappearance of all lesions (nodal short axis <1 cm), partial response (PR) a ≥30% decrease in the sum of target lesion diameters, and progressive disease (PD) a ≥20% increase, new metastatic lesions, or unequivocal progression of non-target lesions; all other cases were classified as stable disease (SD). For bone only metastases assessed according to the PCWG3 2+2 rule, unconfirmed PD was defined as ≥2 new lesions on the first post-treatment scan and confirmed PD defined as ≥2 additional lesions on the subsequent scan, with persistence of the initial lesions.”

Importantly, the authors should discuss whether the use of different imaging modalities and nonuniform imaging follow-up could have influenced the detection of disease progression and the estimation of radiographic progression-free survival.

Response:
Thank you for raising this important point. As expected in a real-world cohort, imaging timing was not uniform and reflected routine clinical practice rather than a protocol-defined imaging schedule. We agree that this heterogeneity could have influenced the timing of progression detection and, consequently, the estimated rPFS. This is reflected in our discussion “Radiographic response rate defined by conventional imaging was lower in our study than in VISION (19% vs. 30%)[2], potentially reflecting differences in baseline disease burden, imaging timing, and inherent variability of imaging assessment in routine practice compared with prospective clinical trial settings.” (page 9).

As rPFS was defined based on conventional imaging, we do not anticipate that the small proportion of patients (6%) who underwent post-treatment PSMA PET/CT substantially influenced rPFS assessment.

The manuscript describes a single-center retrospective study conducted at the University Health Network (UHN), while several sections refer specifically to the Princess Margaret Cancer Centre (PM). The authors should clarify the relationship between these institutions and specify the actual clinical setting in which the study was conducted and the treatment was administered.

Response:
We thank the Reviewer for this inquiry. We clarified the PM is part of UHN (page 3), which also includes our Nuclear Medicine department at Toronto General Hospital.

Added in Introduction” Princess Margaret Cancer Centre (PM) and the affiliated Nuclear Medicine Department in Toronto General Hospital, both within the University Health Network (UHN).” (page 3)

Clarification of the study population and Discrepancy: the manuscript reports different patient numbers across the Results section and supplementary materials, including 77, 60, and 59 patients. Although these values appear to refer to different populations, their definitions and relationships are not sufficiently clear. The authors should clearly explain the difference between these populations. In addition, the radiographic response population includes 64 patients in the main manuscript and 53 patients in Supplementary Table S2. A patient flow diagram illustrating the overall cohort, treatment completion status, and the number of patients evaluable for each outcome would considerably improve the clarity of the manuscript.

Response:
We thank the Reviewer for the opportunity to clarify the study populations.

The overall cohort included 77 patients who initiated treatment, including patients with ongoing treatment at the time of data cut-off. Efficacy outcomes were assessed both in the overall cohort (main manuscript) and separately among patients who had completed or discontinued treatment with available post-treatment assessments (Supplementary Table S2). Among the 77 patients, all 77 had a post-treatment PSA assessment and 64 had at least one post-treatment conventional imaging assessment.

Additional exploratory analyses were performed to evaluate whether inclusion of patients with ongoing treatment, who had incomplete follow up assessment, influenced the observed efficacy estimates. Accordingly, Supplementary Table S2 restricted the analysis to patients who had completed or discontinued treatment, of whom 59 had an available post-treatment PSA assessment and 53 had conventional imaging available for radiographic response assessment.

We have added the following in the Results section:

“All patients had available PSA post treatment therefore evaluable for PSA response.” (page 6)
“Additional exploratory analyses were performed to evaluate whether inclusion of patients with ongoing treatment, who had incomplete follow up and response assessment, may have influenced the response estimates. PSA and radiographic response rates were comparable between the subset of evaluable patients who had completed or discontinued treatment and the overall cohort (Supplementary Materials, Table S2).” (page 7)

An important discrepancy was identified regarding the number of patients achieving a PSA50 response. The main manuscript reports that 31 of 77 patients (40.3%) achieved a PSA50 response, whereas Supplementary Table S4 reports 33 patients with PSA50 response. The authors should carefully review the original dataset and ensure that the number of patients achieving PSA50 response is reported consistently throughout the abstract, Results section, figures, tables, and supplementary materials.

Response:
We thank the Reviewer for catching this. We have reviewed Table S4, the highlighted n numbers were likely a typo, as the rest of the analyses in this table were correct. These two n numbers have been revised in the Supplementary Materials as below, and now matches the corresponding numbers in Table 2.

We have also reviewed all the Tables in the supplementary tables and have not found any additional discrepancies.

Table S4. Impact of selected baseline variables on PSA50 response.

PSA50
No (N=46)	Yes (N=31)	P-value

Several additional discrepancies are present between the main manuscript and the supplementary tables. Some subgroup counts within Supplementary Tables should be rechecked. The authors are therefore encouraged to conduct a comprehensive verification of all reported data, including patient numbers, percentages, subgroup analyses, and statistical results.

Response:
We thank the Reviewer for highlighting this issue. We have reviewed all the Supplementary Tables with the statistician and have revised the tables to ensure consistency. We fixed some formatting errors so the numbers are lined up in the appropriate rows

Minor comments

The abbreviation "GU" in "GU medical oncologists" should be defined at first mention as "genitourinary (GU) medical oncologists".

Response:
We thank the Reviewer for highlighting this. We removed the acronym and described “medical oncologists specialized in Genitourinary malignancies at PM” (page 3)

The supplementary tables are referred to as "Appendix Tables A1–A10" in the main manuscript, whereas the supplementary document labels them as "Tables S1–S10". The numbering should be standardized throughout the manuscript to ensure consistency and facilitate cross-referencing.

Response:
We thank the Reviewer for double checking this. After the initial submission the production team requested us to use Supplementary designation rather than Appendix, so this has already been corrected in the most recent manuscript versions.

The Supplementary Materials section ("The following supporting information can be downloaded at: https://www.mdpi.com/article/doi/s1, Table S1...; Table S2: ; Table S3: ...") appears incomplete.

Response:
We thank the Reviewer for double checking this. The production team has filled out this section which is now complete in the most recent manuscript versions.

Radiopharmaceutical nomenclature should be standardized throughout the manuscript. The notation [¹⁷⁷Lu]Lu-PSMA-617 is recommended for consistency.

Response:
We thank the Reviewer for this suggestion. We have used the suggested notation throughout, including title, abstract (page 1), and throughout the manuscript.

The title and main text should consistently use the term "castration-resistant prostate cancer" rather than "castrate resistant prostate cancer".

Response:
We thank the Reviewer for pointing this out. To align with the recommendation of the previous Reviewer, we have adopted the PCWG4 nomenclature of metastatic androgen pathway modulation-resistant prostate cancer (mAPMR), previously known as mCRPC). This terminology has now been introduced in the title and abstract (pages 1–2), and opening paragraph of the introduction. It has now replaced mCRPC throughout the manuscript.

Reviewer 3 Report

Comments and Suggestions for Authors

Comments to the Authors

This single-centre retrospective study reports the first Canadian real-world experience with [¹⁷⁷Lu]Lu-PSMA-617 after public reimbursement. The manuscript is clearly written, the design is appropriate, and the conclusions are supported by the data. My comments concern internal consistency between sections and some presentation issues.

Major points

Table 1, baseline anemia. Grades 1–3 sum to 77 (100%), yet the hemoglobin range extends to 148 g/L and grade 1 is defined as 100–139 g/L. Please add a "no anemia" category or correct the thresholds. The "9% with Hb <90 g/L" (line 160) cannot be derived from the table; please add it.
Internal vs external patients: 46% + 55% = 101%; please report 45.5%/54.5%.
Treatment completion. Results state that 45% of 60 evaluable patients completed 6 cycles (35% of the whole cohort); the Discussion states "approximately half of the patients". Please specify the denominator and account for patients receiving 5 cycles.
Radiographic progression. Table 2 reports 25 PD + 7 unconfirmed PD, whereas line 198 refers to 45 patients with radiographic progression. Please clarify (best response vs cumulative events).
PSA response denominator. Methods define PSA response "among evaluable patients", but Table 2 uses n=77. Please state that all patients were evaluable.
Prior PSMA radioligand therapy (n=2). Table 1 ("Prior systemic treatment") indicates that 2 patients (2.6%) had previously received PSMA radioligand therapy. This is not mentioned in the Results text (lines 155–158), where the list of prior treatments is limited to ADT, ARPI, docetaxel, cabazitaxel and radium-223, nor is it addressed in the Discussion. Please clarify.
Line 156: "nearly a quarter had also received cabazitaxel or radium-223" — each was 23.4%; the combined proportion would be higher unless fully overlapping. Please rephrase.
SPECT/CT every 2 cycles is described in Methods but no results are reported. Please report or remove.
Internal vs external outcomes. The comparison between internal and externally referred patients is used as a key argument in the Discussion (lines 266–269), where the authors state that externally referred patients achieved comparable outcomes despite higher baseline PSA and greater prior treatment exposure, and use this to support the feasibility of a provincial referral model. However, the corresponding outcome data (PSA response, PSA50 response, PSA-PFS, rPFS and OS by referral source) are reported only as "data not shown" (lines 201–203), and Table S1 contains baseline characteristics only. Since this comparison underpins one of the main implementation conclusions of the paper, the supporting data should be made available. It would also be helpful to report the PET-to-treatment interval separately for the two groups, given that 9 of the 14 delays >90 days occurred among external referrals (lines 165–168), as this would strengthen the discussion on referral pathways and timeliness of access.

Minor points

Nomenclature is inconsistent (177Lutetium-PSMA-617 / 177Lu-PSMA-617 / [¹⁷⁷Lu]Lu-PSMA-617 / Lu-PSMA-617), including title and keywords; please use one form.
"MSH/MMR" vs "MMR/MSI": use dMMR/MSI-H consistently.
"Center"/"Centre"; "a Princess Margaret Cancer Center" (abstract); "Castrate" (title) vs "castration" (text).
Typos: "Prochloperazine" → prochlorperazine; "Publicy"; "statistically significantly"; "did not significant associate"; "177Lu-PSMA-617by".
Methods do not describe how PSA-PFS was calculated.
Consider adding the median PET-to-treatment interval (47 days) to the Abstract, as it is central to the implementation message.

Author Response

Major points

Table 1, baseline anemia. Grades 1–3 sum to 77 (100%), yet the hemoglobin range extends to 148 g/L and grade 1 is defined as 100–139 g/L. Please add a "no anemia" category or correct the thresholds. The "9% with Hb <90 g/L" (line 160) cannot be derived from the table; please add it.

Response:
We thank the Reviewer for highlighting this inconsistency. We have revised baseline anemia descriptions in Table 1 as below (page 4), and also Supplementary Table 1 to be consistent.

We have revised Table 1 for clarity to include no baseline anemia. This was previously included with Grade 1 anemia and now separated out
At UHN our lower limit of normal for hemoglobin in males is defined 140g/L by our institutional lab reference ranges, as such we have defined Grade 1 anemia as 100-139 g/L.
Hb <90 is not defined clearly based on the CTCAE cut offs therefore we have added a separate row to denote in Table 1

Baseline grade of anemia

No anemia (hemoglobin 140 g/L)

Grade 1 anemia (hemoglobin 100 – 139 g/L)

Grade 2 anemia (hemoglobin 80-99 g/L)

Grade 3 anemia (hemoglobin <80 g/L)

5 (6.5)

56 (72.7%)

15 (19.5%)

1 (1.3%)

Baseline hemoglobin <90 g/L

7 (9.1%)

Internal vs external patients: 46% + 55% = 101%; please report 45.5%/54.5%.

Response:
Thank you we have revised these percentages as suggested (page 4)

Treatment completion. Results state that 45% of 60 evaluable patients completed 6 cycles (35% of the whole cohort); the Discussion states "approximately half of the patients". Please specify the denominator and account for patients receiving 5 cycles.

Response:
We thank the Reviewer for highlighting this potential discrepancy. We have clarified “Approximately half of the patients who completed treatment received all six planned cycles” (page 10).

Radiographic progression. Table 2 reports 25 PD + 7 unconfirmed PD, whereas line 198 refers to 45 patients with radiographic progression. Please clarify (best response vs cumulative events).

Response:
We thank the Reviewer for highlighting this apparent discrepancy for clarification. Table 2 is presenting best response including 12 patients who had radiographic response (PR), 20 with stable disease, 7 with PD unconfirmed, and 25 patients with confirmed PD. Among patients who initially did not progress, some did experience radiographic PD subsequently.

At last follow up, 45 patients in total had confirmed radiographic PD. Only 1 out of these 45 patients had confirmed radiographic PD without PSA progression.

To make this important point clearer we have

Added “best” response in the title for Table 2 (page 6)
Added “as best response” when describing radiographic response in text (page 7)
Revised this sentence “At data cut off, 12 patients had ongoing radiographic response, 11 of whom also had ongoing PSA response. Among 45 patients with confirmed radiographic progression at last follow up, all but 1 also had PSA progression.” (page 7)

PSA response denominator. Methods define PSA response "among evaluable patients", but Table 2 uses n=77. Please state that all patients were evaluable.

Response:
Thank you. We added in the Results “All patients had available PSA post treatment therefore evaluable for PSA response.” (page 5)

Prior PSMA radioligand therapy (n=2). Table 1 ("Prior systemic treatment") indicates that 2 patients (2.6%) had previously received PSMA radioligand therapy. This is not mentioned in the Results text (lines 155–158), where the list of prior treatments is limited to ADT, ARPI, docetaxel, cabazitaxel and radium-223, nor is it addressed in the Discussion. Please clarify.

Response:
We thank the Reviewer for highlighting this. We have added in the

Results “Only 2 patients received prior PSMA-targeting radioligand therapy” (page 5)
“Whether retreatment with PSMA-targeted radioligand therapy benefits selected patients is being evaluated in ongoing clinical trials (RE-LuPSMA NCT06288113, VECTRA-01 NCT07611110).” (page 10)

Line 156: "nearly a quarter had also received cabazitaxel or radium-223" — each was 23.4%; the combined proportion would be higher unless fully overlapping. Please rephrase.

Response:
Thank you for pointing this out. We added nearly a quarter “each” to clarify (page 5)

SPECT/CT every 2 cycles is described in Methods but no results are reported. Please report or remove.

Response:
We thank the Reviewer for this suggestion. As we responded to Reviewer 1 unfortunately, we do not have data to present at this time for SPECT / CT we have removed this from Material and Methods (page 3).

Internal vs external outcomes. The comparison between internal and externally referred patients is used as a key argument in the Discussion (lines 266–269), where the authors state that externally referred patients achieved comparable outcomes despite higher baseline PSA and greater prior treatment exposure, and use this to support the feasibility of a provincial referral model. However, the corresponding outcome data (PSA response, PSA50 response, PSA-PFS, rPFS and OS by referral source) are reported only as "data not shown" (lines 201–203), and Table S1 contains baseline characteristics only. Since this comparison underpins one of the main implementation conclusions of the paper, the supporting data should be made available.

Response:
We thank the Reviewer for this excellent suggestion. We have included these exploratory analyses in the Results. On exploratory subgroup analyses, PSA response (62.9% vs 69.0%, p=0.57), PSA50 response (40.0% vs 40.5%, p=0.97), median PSA-PFS (4.2 months [95% CI 3.5-8.3] vs 4.9 months [95% CI 2.7-9.4]), median rPFS (6.7 months [95% 3.7-9.9] vs 6.0 months [95% CI 3.8 – 9.4] and median OS (10.5 months [95% CI 9.4 – NR, not reached] vs NR [95% CI 7.0-NR]) were comparable between internally and externally referred patients respectively. (page 7)

It would also be helpful to report the PET-to-treatment interval separately for the two groups, given that 9 of the 14 delays >90 days occurred among external referrals (lines 165–168), as this would strengthen the discussion on referral pathways and timeliness of access.

Response:
Thank you for this suggestion, we have added this information to the Results section “Median time … was 47 days (range 8-152), and did not differ significantly between internal (41.0 days [range 8.0-152]) and externally referred patients (53.0 days [range 10.0-149.0], p=0.1059.” (page 5)

Minor points

Nomenclature is inconsistent (177Lutetium-PSMA-617 / 177Lu-PSMA-617 / [¹⁷⁷Lu]Lu-PSMA-617 / Lu-PSMA-617), including title and keywords; please use one form.

Response:
We thank the Reviewer for highlighting this important point. To align with the other Reviewer we have adopted [¹⁷⁷Lu]Lu-PSMA-617 throughout the manuscript.

"MSH/MMR" vs "MMR/MSI": use dMMR/MSI-H consistently.

Response:
Thank you. We have revised to use dMMR/MSI-H consistently throughout the manuscript and also the supplementary tables.

"Center"/"Centre"; "a Princess Margaret Cancer Center" (abstract); "Castrate" (title) vs "castration" (text).

Response:
Thank you. We have revised to “Centre” throughout the manuscript. The nomenclature Castrate/castration have been replaced with APMR as previously addressed.

Typos: "Prochloperazine" → prochlorperazine; "Publicy"; "statistically significantly"; "did not significant associate"; "177Lu-PSMA-617by".

Response:
Thank you for pointing these out. We have corrected these to “prochlorperazine” (page 3), “publicly” (page 11), “statistically significant” (page 4), “did not significantly associate” (page 9), “[¹⁷⁷Lu]Lu-PSMA-617 by” (page 2)

Methods do not describe how PSA-PFS was calculated.

Response:
We thank the Reviewer for highlighting this gap in the Methods section. We have added “Time to PSA-PFS, rPFS and OS were calculated from treatment initiation to the date of PSA progression, confirmed radiographic progression or death from any cause respectively. Patients alive without an event were censored at the date of last follow-up.” (page 4).

Consider adding the median PET-to-treatment interval (47 days) to the Abstract, as it is central to the implementation message.

Response:

We thank the Reviewer for this helpful suggestion. We agree that the median 47-day PET-to-treatment interval is relevant to the implementation message. However, as the Abstract is currently at the journal’s word limit, we were unable to add this result without removing other key efficacy or safety findings. The 47-day interval is therefore highlighted in both the Results and Discussion.

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors,

I read your submitted paper regarding the real-life status and results of PSMA treatment, and I found it particularly interesting, enlightening and clinically oriented. I am not stating this just to be “polite” or as a “figure of speech”. The authors appropriately acknowledge the principal limitations of this study, namely its retrospective design, sample size, heterogenous nature, delays in treatment and short follow-up. Below are some suggested points for discussion or improvement.

- The authors state that RECIST 1.1 criteria were applied to soft-tissue lesions and that PCWG3 criteria were applied “whenever possible.” A few words on these criteria would be useful, especially on PCWG3. I am pretty familiar with RECIST but not with PCWG3, and I assume quite a lot of readers will have the same issue.

- The paper does not specify how many patients were assessed in full accordance with these criteria or how response and progression were evaluated in the remaining patients. The first imaging reassessment was performed after 2–4 cycles, and 64/77 patients were evaluable for radiological response. Why were these 13 patients not evaluable for radiological response? How many patients had measurable disease at baseline? How was an overall partial response defined in patients with bone-only or predominantly osseous disease, who made up most of the group? How were the 7 cases described as “unconfirmed progressive disease of bone metastases” handled in the rPFS analysis?

- The patient population includes only those who received at least one dose. It would be interesting to know how many men were referred in total, the number and proportion of patients excluded, and how many deteriorated or died before treatment began. Were there any issues with access to treatment?

- I reckon the presentation of the clinical results of PSMA are a bit “rounded” in terms of phrasing. The results are clinically meaningful, but they are certainly inferior to those reported in clinical trials, and this could be clearly stated. In real life, patients resort to PSMA as a last option: they have poor performance status, lots of comorbidities and very limited physiological reserves, eventually succumbing to their disease.

- Some comments on the cost and cost-effectiveness of the treatment would be useful, since it is a very expensive therapy.

Author Response

Dear authors,

Response:

We sincerely appreciate the Reviewer’s interest and insightful feedback.

We have added the following to the Methodology section to further clarify radiographic response assessment. “In summary for RECIST 1.1, up to five target lesions per metastatic site were recorded as target lesions. Complete response (CR) was defined as disappearance of all lesions (nodal short axis <1 cm), partial response (PR) a ≥30% decrease in the sum of target lesion diameters, and progressive disease (PD) a ≥20% increase, new metastatic lesions, or unequivocal progression of non-target lesions; all other cases were classified as stable disease (SD). For bone only metastases assessed according to the PCWG3 2+2 rule, unconfirmed PD was defined as ≥2 new lesions on the first post-treatment scan and confirmed PD defined as ≥2 additional lesions on the subsequent scan, with persistence of the initial lesions.” (page 3-4)

Response:

Thank you for the opportunity to clarify. At time of data cut off, 13 patients just received their initial cycles of treatment and had not yet completed imaging to assess radiographic response.

How many patients had measurable disease at baseline?

Response:

We did not systemically collect this data in this retrospective study. However given most patients had disease outside the bone (78% lymph nodes, 22% visceral in Table 1), we estimate most patients had measurable disease.

How was an overall partial response defined in patients with bone-only or predominantly osseous disease, who made up most of the group?

Response:

Bone-only or bone predominant disease is assessed according to PCWG3 criteria, primarily to determine progression versus non-progression (PD confirmed or PD unconfirmed), not to assign CR/PR/SD response categories such as RECIST 1.1. PCWG3 is outlined in the Methods (page 3-4). For these patients PSA response is another important metric to assess efficacy.

How were the 7 cases described as “unconfirmed progressive disease of bone metastases” handled in the rPFS analysis?

Response:

Thank you for raising this important to clarify. rPFS was defined as the date of confirmed radiographic progression. We have added “confirmed” in the Methodology section: “Time to PSA-PFS, rPFS and OS were calculated from treatment initiation to the date of PSA progression, confirmed radiographic progression or death from any cause respectively. Patients alive without an event were censored at the date of last follow-up.” (page 4)

Response:

Yes the Reviewer is correct we included patients who had received at least one dose.

We had 42 patients referred from external centers (page 4).
We did not exclude any patients, we add “consecutive” patients to clarify (page 4). However we could only pull patients who received treatment, and we are unfortunately unable to capture at this time how many patients deteriorated or died before treatment began. Our anecdotal experience is that this is a relatively small percentage, as patients with rapidly progressing disease or poor performance status were often not recommended for PSMA RLT.
For our referral volumes, treatment capacity was not a limiting factor at our centre. (Added on page 9)

Response:

Yes we agree with the Reviewer. While the PSA response rates were similar, radiographic response rates and OS were inferior than the VISION trial. We discussed radiographic response rate was lower than in VISION (page 10), and median OS was shorter than in VISION (page 10).

We think the differences in patient population was the main explanation. For example our median PSA was double that in the VISION clinical trial, so these patients had more extensive disease.

In real life, patients resort to PSMA as a last option: they have poor performance status, lots of comorbidities and very limited physiological reserves, eventually succumbing to their disease.

Response:

We agree, and did observe over time as our program opened, initially many patients were getting treatment as clearly a last line therapy (after cabazitaxel). However over time, referrals came in earlier, and included patients who could receive alternative therapies including cabazitaxel We plan to conduct longer follow up and evaluate this important question.

- Some comments on the cost and cost-effectiveness of the treatment would be useful, since it is a very expensive therapy.

Response:

Thank you for this excellent suggestion. We have added a paragraph discussing cost implications for our healthcare system (page 11).

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I thank the authors for their detailed responses and for the revisions made to the manuscript.

The revised version has improved substantially, particularly regarding patient selection, the role of [¹⁸F]FDG PET/CT, the use of conventional imaging for treatment response assessment, and clarification of the institutional setting. Most of my previous methodological comments have been adequately addressed.

However, after reviewing the revised manuscript and supplementary materials, I identified a few remaining issues that should be resolved before publication.

Although the authors state that all supplementary tables have been reviewed, several numerical discrepancies remain between Table 1 and Supplementary Table S1. Since the internal and external cohorts together represent the entire study population, the reported values should be consistent with those presented for the overall cohort.

Furthermore, inconsistencies are still present in Supplementary Table S2, particularly regarding the number of evaluable patients and the reported radiographic response categories.

I therefore recommend a final careful verification of the numerical data, percentages, and corresponding statistical analyses to ensure consistency throughout the manuscript and supplementary materials.

Comments on the Quality of English Language

Minor grammatical and typographical errors should be corrected throughout the manuscript, and all tables, figures, and abbreviations should be checked for consistency.

Author Response

Furthermore, inconsistencies are still present in Supplementary Table S2, particularly regarding the number of evaluable patients and the reported radiographic response categories.

Response:
We thank the Reviewer for your careful review. We requested our statistician to formally review all reported data and corrected additional discrepancies in the following sections.

Table 1 (page 4-5)
- Median eGFR in Table 1 (page 4)
- Somatic / germline mutations
- Prior systemic treatment
- Foot note
Supplementary Table S1
- To keep consistent between Table 1 and Table S1, we added the category “other” under site of disease on PSMA PET in Table S1
- Somatic / germline mutations
- Prior systemic treatment
- Site of lymph node metastases
- Indented row to indicate subcategories in Table 1
Table S2
- PSA and radiographic response outcomes in evaluable patients who completed or discontinued treatment
Tables S6, S7
- Minor changes to p values

We also indented some rows in Tables 1, 2 and Supplementary Table 1 to indicate subcategories for better readability, and hopefully less likely lead to any errors in interpretation.