Visual Abstract
Abstract
177Lu-PSMA-617 therapy has shown high prostate-specific antigen (PSA) response rates in men with metastatic castration-resistant prostate cancer. However, early treatment resistance is common. This LuPIN substudy aimed to determine the prognostic value of posttreatment quantitative PET for PSA progression-free survival (PFS) and overall survival (OS) with 177Lu-PSMA-617 therapy. Methods: Fifty-six men with progressive metastatic castration-resistant prostate cancer were enrolled in the LuPIN trial and received up to 6 doses of 177Lu-PSMA-617 and a radiation sensitizer (NOX66). 68Ga-PSMA-11 and 18F-FDG PET/CT, diagnostic CT, and bone scanning were performed at study entry and exit. Quantitative analysis tracked change in total tumor volume (TTV) and SUV. Univariable and multivariable analyses were conducted to examine the association of change in TTV (continuous and >30%), SUVmax, PSA, and radiographic progression with PSA PFS and OS. Results: All men (37/56) who underwent both screening and posttreatment molecular imaging were analyzed; 70% (26/37) had a PSA response of more than 50%. Median PSA PFS was 8.6 mo, and median OS was 22 mo. Clinical progression had occurred at trial exit in 54% (20/37). In response to treatment, a reduced PSMA SUVmax was demonstrated in 95% (35/37) and a reduced PSMA TTV in 68% (25/37). An increase in PSMA TTV by at least 30% was associated with worse OS (median, 10.2 vs. 23.6 mo; P = 0.002). Change in PSMA SUVmax was not associated with PSA PFS or OS. 18F-FDG SUVmax was reduced in 51% (18/35) and 18F-FDG TTV in 67% (22/35). An increased 18F-FDG SUVmax was associated with worse OS (median, 20.7 vs. 25.7 mo; P < 0.01). An 18F-FDG TTV increase by more than 30% was associated with a short PSA PFS (median, 3.5 vs. 8.6 mo; P < 0.001) but not OS. Both PSA and radiographic progression were associated with shorter OS (median, 14.5 vs. 25.7 mo [P < 0.001] and 12.2 vs. 23.6 mo [P = 0.002]). On multivariable analysis, only increased PSMA TTV and PSA progression remained independently prognostic of OS (hazard ratio, 5.1 [95% CI, 1.5–17.1; P = 0.008] and 3.5 [95% CI, 1.1–10.9; P = 0.03], respectively). Conclusion: Change in quantitative PSMA TTV has strong potential as a prognostic biomarker with 177Lu-PSMA-617 therapy, independent of 18F-FDG PET parameters, PSA, or radiographic progression. Further research into the value of posttreatment PET as an imaging biomarker is warranted.
In the VISION trial, 177Lu-PSMA-617–targeted therapy improved overall survival (OS) and progression-free survival (PFS) in metastatic castration-resistant prostate cancer when compared with the standard of care, and in the TheraP trial it yielded a higher prostate-specific antigen (PSA) response rate and PSA PFS than second-line chemotherapy with cabazitaxel (1,2). However, further work is needed to deepen treatment response and prolong survival. 68Ga-PSMA-11 PET/CT (68Ga-PSMA PET) and 18F-FDG PET/CT (18F-FDG PET) have been used as screening tools in prospective trials to select patients most likely to respond to 177Lu-PSMA-617–targeted treatments (1–4). Less work has been done using molecular imaging to monitor treatment response to 177Lu-PSMA-617 therapy (5–8). Preclinical studies have confirmed that there is considerable interpatient and intrapatient heterogeneity of PSMA expression (9,10). We hypothesized that an increase in uptake or tumor volume on 68Ga-PSMA PET or 18F-FDG PET might have potential as a prognostic biomarker in men being treated with 177Lu-PSMA-617.
In this study, we aimed to determine whether changes in total tumor volume (TTV) and SUV on both 68Ga-PSMA PET and 18F-FDG PET correlate with clinical outcomes in a prospective trial of treatment with 177Lu-PSMA-617.
MATERIALS AND METHODS
This was an imaging substudy of the LuPIN trial. The LuPIN trial is a prospective single center, phase I/II dose escalation and expansion trial of combining 177Lu-PSMA-617 with NOX66. The study enrolled men with metastatic castration-resistant prostate cancer previously treated with both at least 1 line of taxane chemotherapy and an androgen signaling inhibitor. The clinical results have been previously published (11,12). St. Vincent’s Hospital institutional review board approved the study protocol (HREC/17/SVH/19 ACTRN12618001073291), and all participants provided written informed consent.
Screening
Men with progressive metastatic castration-resistant prostate cancer, based on either conventional imaging (CT and bone scanning) or a rising serum concentration of PSA based on Prostate Cancer Working Group 3 criteria (13), were eligible for screening. The men underwent screening with 18F-FDG PET and 68Ga-PSMA PET, bone scanning, and CT of the chest, abdomen, and pelvis. Patients were eligible if they had an SUVmax of more than 15 on 68Ga-PSMA PET at 1 or more sites, an SUVmax of more than 10 at all measurable sites, and no 18F-FDG PET avidity without corresponding PSMA uptake. All men with 68Ga-PSMA PET and 18F-FDG PET at both baseline and after treatment were included in this substudy.
Study Treatment
All men received up to 6 doses of 177Lu-PSMA-617 at 6-wk intervals, with 3 dose-escalated cohorts of NOX66. NOX66 was provided as a dose-appropriate suppository taken from days 1 to 10 after each 177Lu-PSMA-617 injection. All cohorts were administered 7.5 GBq of 177Lu-PSMA-617 on day 1 via a slow intravenous injection. The PSMA-617 precursor (AAA Novartis) was radiolabeled to no-carrier-added 177Lu-chloride according to the manufacturer’s instructions. Quality control tests for radionuclide and radiochemical purity were performed using high-pressure liquid chromatography and thin-layer chromatography. NOX66 suppositories were administered at 400-, 800-, and 1,200-mg doses per a dose escalation protocol (11).
Imaging Procedures and Acquisition
68Ga-PSMA PET and 18F-FDG PET were performed at baseline (screening) and after treatment (6 wk after completing all 6 cycles or when treatment ceased earlier because of clinical progression). 68Ga-HBEDD-CC PSMA-11 was produced on-site in a manner compliant with good-laboratory-practice procedures using a TRASIS automated radiopharmacy cassette. 18F-FDG was produced off-site commercially under good-manufacturing-practice–compliant conditions. Radiopharmacy quality control was undertaken using a high-pressure liquid chromatography method. Patients were injected with a 2.0 MBq/kg dose of 68GaPSMA-11 and a 3.5 MBq/kg dose of 18F-FDG, with matched imaging parameters (dose, time after injection, and imaging protocols) for each patient. All PET/CT imaging was undertaken using a Phillips Ingenuity time-of-flight PET/64-slice CT scanner. An unenhanced low-dose CT scan was performed 60 min after tracer injection. Immediately after CT scanning, a whole-body PET scan was acquired for 2 min per bed position. The emission data were corrected for randoms, scatter, and decay.
Diagnostic contrast-enhanced CT of the chest, abdomen, and pelvis, and a whole-body bone scan, were performed at baseline and after treatment.
Imaging Analysis
All 68Ga-PSMA and 18F-FDG PET scans (screening and after treatment) were analyzed semiquantitatively by a nuclear medicine physician using MIM Software and a standardized semiautomated workflow to delineate regions of interest with a minimum SUV cutoff of 3 for 68Ga-PSMA PET and an SUV cutoff equal to the blood pool mean intensity plus 1.5 SDs for 18F-FDG PET. All lesions identified quantitatively were manually reviewed and physiologic uptake or scatter removed. Whole-body quantitation derived total metabolic tumor volume, SUVmax, and SUVmean for both 18F-FDG PET and 68Ga-PSMA PET (MIM Software) (14). A nuclear medicine physician visually assessed both the quantified and the nonquantified PET images to identify potential sites of 18F-FDG PET–positive/68Ga-PSMA PET–negative progressive disease between the screening and posttreatment scans.
Statistical Analyses
We measured PSA declines from baseline (absolute and ≥50%) at any time point, PSA PFS as defined by Prostate Cancer Working Group 3 criteria, radiographic progression defined by RECIST 1.1 and Prostate Cancer Working Group 3 criteria, and OS (13,15). Time-to-event endpoints (PSA PFS and OS) were defined as the interval from the date of enrolment to the event date, or the date last known to be event-free (at which point the observation was censored).
A 2-sided exact binomial 95% CI was calculated for PSA response rates. The Kaplan–Meier method was used to characterize time-to-event endpoints and estimate medians (presented with 95% CIs). We correlated changes in TTV, SUVmax, and SUVmean for 68Ga-PSMA PET and 18F-FDG PET with time-to-event outcomes, using univariable and multivariable Cox proportional-hazards regression models. P values below 5% were considered significant. Analyses were performed using R (version 4.0.5) and SPSS (version 25).
RESULTS
Patient Characteristics
Patient characteristics are summarized in Table 1. Thirty-seven of 56 (66%) men on the LuPIN trial had both baseline screening and posttreatment imaging (6 wk after completion of 6 cycles of treatment, or earlier if the trial was exited because of clinical progression). Of these, 68% (25/37) had posttreatment imaging after completing all 6 cycles of 177Lu-PSMA-617 plus NOX66, 3% (1/37) after 5 cycles, 14% (5/37) after 4 cycles, 14% (5/37) after 3 cycles, and 3% (1/37) after 2 cycles. Nineteen of 56 (34%) did not have exit imaging because of illness (12/19), travel restrictions (3/19), or unknown reasons (3/19).
Patient Characteristics
Clinical Outcomes
The median reduction in PSA was 77% (interquartile range, 34%–92%), and 70% (26/37) of patients had a PSA response of more than 50%. With a median follow-up of 26 mo, the median PSA PFS was 8.6 mo (95% CI, 5.6–11.6) and the median OS was 22 mo (95% CI, 18.6–25.6). PSA or clinical progression had occurred in 54% (20/37) at the time of exit imaging, whereas 46% (17/37) had no PSA progression after the full 6 cycles of treatment. At exit, 92% (34/37) had conventional imaging (CT and bone scanning). Radiographic progression at the time of exit imaging was identified in 18% (6/34). On univariable analysis, PSA progression and radiographic progression at trial exit were also associated with significantly worse PSA PFS and OS (Table 2).
Univariable Cox Regression Analysis for Association with PSA PFS and OS
68Ga-PSMA PET Quantitation
Quantitative 68Ga-PSMA PET SUVmax and SUVmean were reduced in 95% of men independently of whether they had PSA progression at the time of exit imaging (absolute change in PSMA SUVmax: median, −26 [interquartile range, −40 to −13]; absolute change in PSMA SUVmean: median, −3 [interquartile range, −5 to −2]). There was no correlation between an increase in PSMA SUVmax and either PSA PFS or OS (Table 2).
The median change in PSMA TTV was −0.64 liters (interquartile range, −0.29 to +0.07). PSMA TTV was increased in 32% (12/37). Any increase in PSMA TTV was associated with shorter PSA PFS (hazard ratio [HR], 2.9 [95% CI, 1.4–6.1]; P = 0.01) and OS (median, 12.2 vs. 25.5 mo; HR, 6.2 [95% CI, 2.0–19.2]; P < 0.01) (Fig. 1). An increase in PSMA TTV by at least 30% was significantly associated with worse OS (HR, 6.0 [95% CI, 1.9–19.2]; P = 0.002), whereas association with PSA PFS was not significant (Fig. 2).
(A and B) Quantitative analysis for patient with reduced PSMA TTV between baseline (A) and after treatment (B). (C and D) Quantitative analysis for progressing patient at baseline (C) and after treatment (D). In patients with high-volume disease, it can be difficult to visually identify extent of volume change. In this second case, volume increase is 25%.
Kaplan–Meier curves for OS stratified by increase in PSMA TTV ≥ 30% from baseline (A), increase in 18F-FDG TTV ≥ 30% from baseline (B), increase in PSMA SUVmax from baseline (C), and increase in 18F-FDG SUVmax from baseline (D).
All 12 patients with an increasing PSMA TTV had an SUVmax of more than 15 at trial exit (above trial entry criteria), compared with 52% (13/25) of those with a reduced PSMA TTV.
18F-FDG PET Quantitation
Analysis of screening and posttreatment 18F-FDG PET demonstrated that 51% (18/35) had a reduced 18F-FDG SUVmax (median absolute change, 0.1; interquartile range, −4 to +1) and that 66% (23/35) had a reduced 18F-FDG SUVmean (median absolute change, −0.4; interquartile range, −1 to +0.4) in response to treatment. An increase in 18F-FDG SUVmax was associated with worse PSA PFS (HR, 3.0 [95% CI, 1.4–6.4]; P = 0.01) and OS (HR, 3.0 [95% CI, 1.2–7.3]; P = 0.02).
The median change in 18F-FDG TTV was −0.01 liters (interquartile range, −0.05 to +0.02). 18F-FDG TTV was increased in 37% (13/35) at trial exit. Any increase in 18F-FDG TTV was associated with worse PSA PFS (HR, 3.1 [95% CI, 1.4–6.8]; P = 0.005) and OS (median, 16.2 vs. 23.1 mo; HR, 2.7 [95% CI, 1.1–6.2]; P = 0.02). An increase in 18F-FDG TTV by at least 30% was significantly associated with worse PSA PFS (HR, 2.7 [95% CI, 1.2–6.0]; P = 0.01) but was not significantly associated with OS (Fig. 2).
No 18F-FDG–positive/68Ga-PSMA PET–negative progressive sites were identified in this cohort at the time of exit imaging. One patient had a significant increase in 18F-FDG–avid volume at 1 site, whereas PSMA tumor volume and SUVmax were reduced (Fig. 3).
Quantitative analysis of baseline and posttreatment PSMA and 18F-FDG PET images from same patient: baseline 68Ga-PSMA PET (A), posttreatment 68Ga-PSMA PET (B), baseline 18F-FDG PET (C), and posttreatment 18F-FDG PET (D). Arrows indicate lesion in iliac bone that reduced in volume and intensity on 68Ga-PSMA PET but increased in volume and intensity of 18F-FDG PET.
Molecular Response Patterns and Patient Outcomes
Multivariable analysis including change in PSMA TTV, 18F-FDG TTV, and 18F-FDG SUVmax, as well as PSA progression and radiographic progression, found that only PSMA TTV and PSA progression remained independently prognostic of OS (HR, 5.1 [95% CI, 1.5–17.1; P = 0.008] and 3.5 [95% CI, 1.1–10.9; P = 0.03], respectively) (Table 3).
Multivariable Cox Regression Analysis for Association of Clinical and Imaging Parameters with OS
DISCUSSION
This study has found that increasing TTV on posttreatment 68Ga-PSMA PET identifies early disease progression and shorter OS independently of PSA, raising its potential for use as a prognostic biomarker. Metastatic castration-resistant prostate cancer is characterized by phenotypic and molecular heterogeneity, with marked PSMA heterogeneity previously demonstrated at both an imaging and a cellular level (9,10). Although the VISION and TheraP trials have found high treatment responses and improved quality-of-life parameters, the duration of treatment responses with 177Lu-PSMA-617 remains limited. Identifying effective predictive and prognostic biomarkers is critical to deepening and prolonging responses to PSMA-targeted treatments with appropriate combinations and judicious treatment sequencing.
A second key finding in this study is that in contrast to 18F-FDG PET, reduced PSMA SUVmax or SUVmean occurred in almost all patients in response to 177Lu-PSMA-617 therapy and was not predictive of either treatment response or OS. This lack of correlation between change in PSMA SUVmax or SUVmean and treatment outcomes has previously been shown. Kurth et al. found that PSMA intensity decreased in both clinically responding and progressing patients (7,16). Grubmüller et al. also found no correlation between change in whole-body PSMA SUVmean and OS in an analysis of posttreatment 68Ga-PSMA PET after 177Lu-PSMA-617 therapy. This lack of prognostic value of change in PSMA SUVmean and SUVmax with PSMA-targeted therapy is not unexpected. 177Lu-PSMA-617 preferentially targets highly PSMA-expressing cells, leading to persistent populations of low-PSMA-expression disease that may be less responsive to treatment. The lack of predictive or prognostic value for change in PSMA SUVmean and SUVmax is important to highlight, as we intuitively use reduction in intensity (18F-FDG PET) to denote treatment response with systemic therapy (17). We need to think differently when developing 68Ga-PSMA PET response criteria for PSMA-targeted therapy.
Increasing PSMA TTV was an independent predictor of PSA PFS and OS in this study. Similar to Gafita et al. and Grubmüller et al., we confirmed that an increase in quantitative PSMA TTV is a poor prognostic factor for OS (6,7). An increase in PSMA TTV by at least 30% was associated with poor OS, supporting the inclusion of this metric in the 68Ga-PSMA PET progression criteria (18). However, accurate assessment of change in TTV visually can be difficult, especially in high-volume disease, and quantitative PET analysis may become an important tool in PSMA-targeted therapy.
We found that patients with an increase in PSMA TTV at trial exit had a PSMA SUVmax of more than 15, significantly higher than in patients without progressive disease, and above a range at which PSMA-targeted therapy is expected to be effective. This finding may indicate that radiation resistance is an important mechanism of treatment failure. Further evaluation in conjunction with genetic analysis may help identify optimal treatment combinations in patients who currently have a limited treatment response to 177Lu-PSMA-617 alone.
18F-FDG PET was undertaken both at screening and at trial exit, with 18F-FDG PET screening parameters previously shown to be predictive of OS in this study cohort and other 177Lu-PSMA-617 trials (1,12,19). Although we found that an increase in 18F-FDG SUVmax and 18F-FDG TTV was associated with poor OS in univariable analysis, they did not remain significant on multivariate analysis. Further, the incidence of discordant progressive lesions (18F-FDG PET–positive/68Ga-PSMA PET–negative) was low, with only 1 patient having a significant increase in 18F-FDG–avid volume at 1 site, whereas PSMA TTV was reduced.
RECIST progression is the standard of care for identifying progressive disease on imaging and has a strong correlation with OS in prostate cancer (20). However, RECIST progression was less prognostic than change in either PSMA TTV or PSA progression at study exit. The promising prognostic value for molecular imaging parameters suggests that more work needs to be done validating PET for treatment response in metastatic castration-resistant prostate cancer, potentially as an alternative to CT and bone scanning currently used in Prostate Cancer Working Group 3 criteria.
This study had several limitations. Its small sample size made it purely exploratory, and the findings will need to be validated in larger trials. The sample size also did not allow for a multivariable analysis incorporating other known prognostic factors.
Additionally, selection of patients for this analysis was biased toward those well enough to complete posttreatment imaging, explaining the higher rate of at least 50% PSA decline and longer OS in this subset of patients than previously published for the trial. The interval between screening and posttreatment imaging was variable, with 32% exiting the trial early because of clinical progression.
Finally, PET quantitation software remains of limited availability and time-intensive to achieve accurate results. Further automation in quantitation is required to minimize the time required to derive reproducible results, in addition to harmonization and validation of quantitative methods. Nevertheless, the prognostic value of quantified PSMA TTV in this study suggests that investment in PET quantitation will yield significant clinical benefit.
CONCLUSION
Change in quantitative PSMA TTV has strong potential as a prognostic biomarker with 177Lu-PSMA-617 therapy, independently of 18F-FDG parameters, PSA, or radiographic progression. Further research into the value of posttreatment PET as an imaging biomarker is warranted.
DISCLOSURE
This investigator-initiated study was sponsored by St. Vincent’s Hospital Sydney and supported by a Cancer Institute NSW prostate translational research grant. Noxopharm Limited provided funding for drug and PET scans, AAA/Novartis provided PSMA-617 ligand. Louise Emmett has an advisory role with Noxopharm, receives trial support from Novartis and Astellas, and receives grant funding support from St. Vincent’s Clinic Foundation. Anthony Joshua has an advisory role with Noxopharm Limited and receives institutional funding from Novartis. Peter Wilson and Remy Niman are salaried employees of MIM Software, Inc. No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: What are the imaging findings on 68Ga-PSMA PET and 18F-FDG PET after 177Lu-PSMA-617 therapy, and do changes in tumor volume, SUVmax, or SUVmean correlate with clinical outcomes?
PERTINENT FINDINGS: In this LuPIN substudy, any increase in PSMA TTV and PSA progression at study exit were independently prognostic of OS.
IMPLICATIONS FOR PATIENT CARE: Change in tumor volume on 68Ga-PSMA PET after 177Lu-PSMA-617 therapy provides information for clinicians on patient survival and may help them make clinical decisions on timing and type of next treatments.
ACKNOWLEDGMENTS
We thank the patients, and the clinical trials teams at the Department of Theranostics and Nuclear Medicine and the Kinghorn Cancer Centre, St. Vincent’s Hospital, for their support.
Footnotes
Published online Jun. 23, 2022.
- © 2023 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication March 5, 2022.
- Revision received June 13, 2022.