Visual Abstract
Abstract
A single-institution prospective pilot clinical trial was performed to demonstrate the feasibility of combining [177Lu]Lu-PSMA-617 radiopharmaceutical therapy (RPT) with stereotactic body radiotherapy (SBRT) for the treatment of oligometastatic castration-sensitive prostate cancer. Methods: Six patients with 9 prostate-specific membrane antigen (PSMA)–positive oligometastases received 2 cycles of [177Lu]Lu-PSMA-617 RPT followed by SBRT. After the first intravenous infusion of [177Lu]Lu-PSMA-617 (7.46 ± 0.15 GBq), patients underwent SPECT/CT at 3.2 ± 0.5, 23.9 ± 0.4, and 87.4 ± 12.0 h. Voxel-based dosimetry was performed with calibration factors (11.7 counts per second/MBq) and recovery coefficients derived from in-house phantom experiments. Lesions were segmented on baseline PSMA PET/CT (50% SUVmax). After a second cycle of [177Lu]Lu-PSMA-617 (44 ± 3 d; 7.50 ± 0.10 GBq) and an interim PSMA PET/CT scan, SBRT (27 Gy in 3 fractions) was delivered to all PSMA-avid oligometastatic sites, followed by post-PSMA PET/CT. RPT and SBRT voxelwise dose maps were scaled (α/β = 3 Gy; repair half-time, 1.5 h) to calculate the biologically effective dose (BED). Results: All patients completed the combination therapy without complications. No grade 3+ toxicities were noted. The median of the lesion SUVmax as measured on PSMA PET was 16.8 (interquartile range [IQR], 11.6) (baseline), 6.2 (IQR, 2.7) (interim), and 2.9 (IQR, 1.4) (post). PET-derived lesion volumes were 0.4–1.7 cm3. The median lesion-absorbed dose (AD) from the first cycle of [177Lu]Lu-PSMA-617 RPT (ADRPT) was 27.7 Gy (range, 8.3–58.2 Gy; corresponding to 3.7 Gy/GBq, range, 1.1–7.7 Gy/GBq), whereas the median lesion AD from SBRT was 28.1 Gy (range, 26.7–28.8 Gy). Spearman rank correlation, ρ, was 0.90 between the baseline lesion PET SUVmax and SPECT SUVmax (P = 0.005), 0.74 (P = 0.046) between the baseline PET SUVmax and the lesion ADRPT, and −0.81 (P = 0.022) between the lesion ADRPT and the percent change in PET SUVmax (baseline to interim). The median for the lesion BED from RPT and SBRT was 159 Gy (range, 124–219 Gy). ρ between the BED from RPT and SBRT and the percent change in PET SUVmax (baseline to post) was −0.88 (P = 0.007). Two cycles of [177Lu]Lu-PSMA-617 RPT contributed approximately 40% to the maximum BED from RPT and SBRT. Conclusion: Lesional dosimetry in patients with oligometastatic castration-sensitive prostate cancer undergoing [177Lu]Lu-PSMA-617 RPT followed by SBRT is feasible. Combined RPT and SBRT may provide an efficient method to maximize the delivery of meaningful doses to oligometastatic disease while addressing potential microscopic disease reservoirs and limiting the dose exposure to normal tissues.
Prostate cancer is the most common and the second most lethal cancer among men in the United States, with 288,300 estimated new diagnoses and 34,700 estimated deaths in 2023 (1). An oligometastatic clinical state has been proposed as an intermediate point between local and widespread disease (2). Patients with more limited metastatic burden might thus benefit from aggressive metastasis-selective therapy. In a recent pooled analysis of prospective randomized phase II trials in oligometastatic castration-sensitive prostate cancer (omCSPC), median progression-free survival was prolonged with metastasis-selective therapy compared with observation (3).
Stereotactic body radiotherapy (SBRT) is an effective way of delivering high radiation doses for patients with oligometastases. It has been widely adopted (4) as a generally well-tolerated treatment strategy with high local control rates (5). Distant metastasis-free survival rates are, however, modest (6), as patients often develop metastatic cancer outside of the treated area (7). Targeted radiopharmaceutical therapy (RPT) with [177Lu]Lu-PSMA-617 (Pluvicto; Novartis AG) is an efficacious systemic treatment, delivering cytotoxic β-radiation to cells expressing prostate-specific membrane antigen (PSMA) without causing excessive normal-tissue toxicity (8–10). Outcomes for omCSPC may be improved by combining SBRT for focal therapy of bulk tumors with [177Lu]Lu-PSMA-617 RPT to address microscopic disease reservoirs (11). Because of nonoverlapping toxicities, these treatment strategies allow escalation of the combined absorbed dose (AD) of radiation in the tumor and may act synergistically (12,13). Because both use ionizing radiation, a quantitative metric combining the radiation ADs from each treatment into a biologically effective dose (BED) can be calculated, which in turn can be used in models to predict the response and toxicity (14,15). Combination therapies with radiopharmaceuticals are becoming a new paradigm (11), with 2 currently ongoing clinical studies using external-beam radiotherapy (EBRT) and [177Lu]Lu-PSMA-617: PROQURE-1 phase I trial (16) and LUNAR phase II trial (17). The present study was performed to evaluate the feasibility of performing composite lesional dosimetry in PSMA-avid omCSPC patients undergoing theranostic [177Lu]Lu-PSMA-617 RPT followed by SBRT.
MATERIALS AND METHODS
Patient Selection
The pilot clinical trial was approved by Memorial Sloan Kettering Cancer Center’s Institutional Review Board (protocol 21-158; NCT05079698), and all patients gave written informed consent regarding the examination and use of anonymous data for research and publication purposes. Inclusion and exclusion criteria are available in the supplemental materials (available at http://jnm.snmjournals.org). Briefly, patients with a biopsy-confirmed adenocarcinoma of the prostate, a prostate-specific antigen (PSA) level between 0.5 and 50 ng/mL, and a primary tumor previously (>2 y) treated with surgery or definitive radiation were considered. The oligometastatic patient population was defined as oligorecurrence after definite prostate-directed therapy with 1–3 discrete sites of gross metastatic disease.
PSMA PET/CT Imaging
PSMA PET/CT was used to identify patients, gauge response, and track efficacy (18–20). PSMA PET/CT scans were performed at baseline, after [177Lu]Lu-PSMA-617 RPT (interim), and after SBRT (post). Patients were injected with 208 ± 38 MBq (range, 167–266 MBq) of 68Ga-PSMA-11 or 332 ± 34 MBq (range, 266–366 MBq) of 18F-DCFPyL and imaged 71 ± 9 min after administration (range, 60–95 min). Whole-body (vertex of skull to proximal thigh) images were acquired on a GE Healthcare Discovery 710 PET/CT scanner. 18F-DCFPyL was previously reported to result in an approximately 18% higher lesion SUVmax (21); however, in that study, 18F-DCFPyL PET images were acquired 2 h after injection versus 1 h after injection for 68Ga-PSMA-11 PET. A separate study indicated that the 18F-DCFPyL lesion SUVmax increased by approximately 19% between 1 and 2 h (22). The CT images were acquired at 120 kVp, 56 mAs, and a 3.75-mm slice thickness. All PET emission data were corrected for attenuation, scatter, and random events and iteratively reconstructed into a 128 × 128 × 47 matrix (voxel dimensions, 5.47 × 5.47 × 3.27 mm) using an ordered-subset expectation maximization algorithm (2 iterations, 16 subsets) incorporating time-of-flight and point-spread-function modeling. A gaussian postprocessing filter of 6.4 mm in full width at half maximum was applied. All screening and posttreatment PSMA scans were reviewed by an experienced nuclear medicine physician, and lesions were scored using the structured PSMA reporting and data system, version 1.99, radiographic criteria (23). Treatable oligometastases were lesions scored as either 4 or 5 on this system.
[177Lu]Lu-PSMA-617 SPECT/CT Imaging
After the first intravenous infusion of [177Lu]Lu-PSMA-617 (7.46 ± 0.15 GBq), patients were serially imaged on a Symbia Intevo Bold (Siemens Healthineers AG) γ-camera equipped with a medium-energy collimator and 9.5-mm-thick crystals. Imaging consisted of a whole-body planar scintigraphy (anterior and posterior; scan speed, 10 cm/min) followed by a single-field-of-view SPECT/CT scan. SPECT/CT start times were 3.2 ± 0.5, 23.9 ± 0.4, and 87.4 ± 12.0 h after infusion, with acquisition parameters as follows: 180° elliptic arc, step-and-shoot mode with 64 stops, 20 s/stop, 20% acquisition window at 208 keV with an adjacent 10% scatter windows, and a 128 × 128 × 79 matrix (4.8 × 4.8 × 4.8 mm voxels). SPECT image reconstruction was performed in HybridRecon version 3.0 (Hermes Medical Solutions) and incorporated motion, attenuation, resolution recovery, and scatter corrections. An ordered-subset expectation maximization algorithm (48 iterations, 1 subset) without a postprocessing filter was used for image reconstruction (24,25). For SPECT quantification, a calibration factor (11.7 counts per minute/MBq) was derived from in-house experiments with a 177Lu-filled American College of Radiology SPECT phantom without inserts. After the second [177Lu]Lu-PSMA-617 administration (7.50 ± 0.10 GBq; 44 ± 3 d after the beginning of cycle 1), whole-body planar scintigraphy (with same acquisition parameters as used during the first cycle) was performed on the day of injection only, with no SPECT imaging.
SBRT
Patients underwent SBRT on a Varian TrueBeam linear accelerator to all PSMA-avid oligometastatic sites (prescription of 27 Gy in 3 fractions; institutional standard for extracranial lesions) at 5 ± 1 wk after the second [177Lu]Lu-PSMA-617 cycle and after simulation and radiation planning (Eclipse version 16; Varian Medical Systems Inc.) according to Memorial Sloan Kettering Cancer Center’s Department of Radiation Oncology guidelines. Before the treatment, patients were immobilized in a reproducible position in a custom mold. A CT scan was acquired in the treatment-planning position (120 kVp, 56 mAs, and 3-mm slice thickness) on a Philips Big Bore CT simulator. The dose was prescribed to the 100% isodose line, which completely encompassed the planning tumor volume. SBRT was delivered with 6× flattening filter-free photon beam profiles. Hot spots were limited to less than 110% of the prescription dose. Normal tissues and lesions were contoured on the treatment-planning CT to determine dose–volume histograms. The Digital Imaging and Communications in Medicine dose, plans, structures, and CT data were subsequently imported from Eclipse version 16.
Dosimetry
Voxel-based dosimetry was performed in Velocity version 4.2 Development Build (Varian Medical Systems Inc.). Lesions were segmented on the coregistered PSMA PET/CT (50% SUVmax threshold) and subsequently copied to the [177Lu]Lu-PSMA-617 SPECT/CT. AD calculations were based on the time–activity data derived from each patient’s set of 3 SPECT/CT images. Application of partial-volume correction was based on recovery coefficients (RCs) derived from in-house phantom experiments with a standard National Electrical Manufacturers Association Image Quality phantom with 6 spheric inserts (diameter, 10–37 mm). The phantom was imaged and analyzed using acquisition settings and reconstruction parameters identical to those used for the clinical protocol. Volumes of interest (VOIs) were drawn on the CT image and subsequently copied to the SPECT image. The fit to our RC versus sphere diameter data was
Eq. 1where sd is the sphere diameter in millimeters. Comparison with RC obtained by other groups is presented in Supplemental Figure 1 (26,27). Voxel-based dosimetry was performed with the ACUROS Molecular Radiotherapy algorithm within Velocity, which uses a voxel-based fitting method that estimates a fitting function (sum of exponentials) for each voxel by automatically selecting the most appropriate model among a predetermined set via Akaike information criterion (28).
As a cross-check, the area-under-the-curve (MBq·h/mL) and time-integrated activity coefficients for lesions were also calculated according to the scheme defined by the committee on MIRD (29). Cumulated activity within the lesion VOI was determined by a trapezoid method between the end of the infusion and the third SPECT scan and a monoexponential model without residual activity thereafter. Lesion ADs were subsequently calculated using the sphere model in OLINDA/EXM version 1.1 (Hermes Medical Solutions). A third method using multiple concentric oversized VOIs to determine the activity of a source was also investigated (10,30). Unlike the first 2 approaches, the latter method does not require the use of RC.
BED scaling of both RPT and SBRT dose maps was performed in Velocity. BED is defined as
Eq. 2where D is the total dose and RE is the relative effectiveness. For treatments with fractionated doses (i.e., SBRT), D and RE are calculated as
Eq. 3
Eq. 4where n and d are the number of fractions (3) and fraction size (9 Gy), respectively. The α/β ratio was set to 3 Gy. For treatments with an exponentially decaying source integrated to infinity (i.e., [177Lu]Lu-PSMA-617 RPT) and a monoexponential dose-rate function, RE is defined as (14)
Eq. 5
Eq. 6where R0 is the initial dose rate (Gy/h), λ is the effective decay rate, and μ is the repair rate (set to 0.462 h−1, corresponding to a commonly used repair half-time of 1.5 h). In this special case, λ/(μ + λ) is the Lea–Catcheside time factor. BED-scaled dose maps were subsequently aligned (deformable multipass coregistration) and summed into a single resampled dose volume, from which dose–volume histograms were extracted. Note that the equieffective dose, EQD2α/β (Gy), where 2 refers to 2-Gy daily fractions, can be calculated as
Eq. 7
For an α/β ratio of 3, an EQD2α/β of 3 is equal to 0.6 × BED (BED is mathematically the same as a dose delivered in infinitely small 0-Gy fractions).
Statistical Analysis
The Spearman rank-order correlation coefficient, ρ, was calculated to evaluate the strength of association between various investigated metrics. Metrics between different time points (baseline, post-RPT, post-SBRT) were compared with the Wilcoxon signed-rank test (2-tailed). Interquartile range (IQR) was used as a measure of statistical dispersion. A P value of less than 0.05 indicated statistical significance. The statistical analysis was performed in MATLAB R2020b (MathWorks Inc.) with Statistics and Machine Learning Toolbox version 12.0.
RESULTS
Six patients (median age, 74 y; range, 51–78 y) with PSMA-positive metachronous oligometastases were included in the analysis (Table 1). The first cycle of [177Lu]Lu-PSMA-617 RPT was administered between December 2021 and September 2022. The time after the initial definite prostate-directed therapy was 7 y (range, 4–24 y). Baseline PSA levels were 2.0 ng/mL (IQR, 1.2 ng/mL). All patients underwent hematologic safety and adverse event monitoring for at least 4 wk after both cycles of [177Lu]Lu-PSMA-617 RPT. Two patients experienced grade 2 toxicities after the combined treatment: transient anemia and hyperbilirubinemia (the latter is probably unrelated; no patients had liver metastases). No grade 3+ toxicities were noted. Nine PSMA-positive lesions were identified by the nuclear medicine physician. Lesion volume as measured on baseline PSMA PET/CT was 0.8 cm3 (range, 0.4–1.7 cm3), resulting in a median SPECT RC of 0.20 (IQR, 0.01; range, 0.16–0.30). Median of the lesion SUVmax as measured on PSMA PET was 16.8 (IQR, 11.6), 6.2 (IQR, 2.7), and 2.9 (IQR, 1.4) at baseline, interim (i.e., after completion of [177Lu]Lu-PSMA-617 RPT but before SBRT), and post (i.e., after completion of SBRT), respectively. Percentage change (%Δ) in SUVmax between baseline and interim PSMA PET (calculated as 100% × (interim SUVmax – baseline SUVmax)/baseline SUVmax, that is, %Δ1) was −65% (range, −82% to 44%; P < 0.05). Corresponding %Δ between interim and post-PSMA PET (i.e., %Δ2) was −43% (range, −74% to 31%; significance not reached), whereas %Δ between baseline and post-PSMA PET (i.e., %Δtotal) was −74% (range, −91% to −30%; P < 0.05). The %Δ1 in the PSMA PET–derived maximum tumor-to-liver ratio (calculated as lesion SUVmax/liver SUVmean) was −59% (P < 0.05), whereas the corresponding %Δ2 and %Δtotal for the maximum tumor-to-liver ratios were −49% and −77%, respectively (P < 0.05). The median %Δ in PSA between baseline and 3 mo after the combined therapy was −83% (range, −97% to 58%; significance not reached because of paucity of data), closely resembling the relative decrease in PSMA SUVmax.
Summary of Patient Characteristics and Imaging Results
Seven of 9 lesions remained PSMA-avid after [177Lu]Lu-PSMA-617 treatment, as assessed by a lesion uptake greater than the mean liver uptake on the interim PSMA PET/CT scan. Liver SUV and parotid SUV were not significantly different among the 3 PSMA PET scans (liver SUV = 5.4, 4.6, and 5.0, respectively; parotid SUV = 12.5, 12.4, and 10.5, respectively). Three of 9 lesions (33%; lesions 1, 2, and 4) were visible on the standard-of-care 99mTc-methylene diphosphonate planar bone scintigraphy (acquired 2.5–3 h after 774 ± 51 MBq administration of 99mTc-methylene diphosphonate) at both baseline and post-SBRT imaging. The remaining lesions were lymph nodes (n = 5) or visceral (n = 1). No additional lesions were noted on the 99mTc-methylene diphosphonate scans that were not visualized by PSMA PET/CT.
In patient 1, the 2 lesions could not be visualized on the single-bed SPECT scan (380-cm axial field of view); therefore, dosimetric calculations were performed for only 1 (n = 8 lesions in total). Median of the lesion SUVmax as measured on [177Lu]Lu-PSMA-617 SPECT was 3.7 (IQR, 1.5), 9.4 (IQR, 7.2), and 7.3 (IQR, 4.5) on the first, second, and third SPECT, respectively. The highest lesion SUVmax was reached on either the second or the third SPECT scan. The salivary glands were not within the SPECT field of view for 5 of 6 patients.
The summary of dosimetry results is presented in Table 2. The mean lesion AD from the first cycle of [177Lu]Lu-PSMA-617 RPT (ADRPT; calculated from voxels within the tumor volume as defined by the 50% threshold on PSMA PET) was 27.7 Gy (IQR, 17.5 Gy; range, 8.3–58.2 Gy), translating into 3.7 Gy/GBq (IQR, 2.4 Gy/GBq; range, 1.1–7.7 Gy/GBq). As β electrons emitted from 177Lu have an approximate 0.6-mm range in soft tissue, most of the energy will be absorbed within 1 cm3 lesions. The mean BED from RPT was 28.2 Gy (IQR, 18.5 Gy; range, 8.4–59.9 Gy; median RE from RPT, 1.02). The medians of the maximum ADRPT and maximum BED from RPT were 37.5 Gy (IQR, 24.2 Gy) and 38.5 Gy (IQR, 25.4 Gy), respectively.
Summary of Dosimetry Results
The median ADRPT when recalculated using the MIRD formalism, and the sphere model in OLINDA was 32.7 Gy (IQR, 20.4 Gy; range, 7.7–70.5 Gy), translating into 4.3 Gy/GBq (IQR, 2.8 Gy/GBq; range, 1.0–9.6 Gy/GBq). The corresponding median ADRPT calculated with a multiple concentric oversized VOI method was 27.3 Gy (IQR, 10.2 Gy; range, 16.1–42.9 Gy), translating into 4.1 Gy/GBq (IQR, 1.2 Gy/GBq; range, 2.1–5.8 Gy/GBq). ADRPT values calculated with both of these methods were strongly correlated with ADRPT derived in Velocity (Spearman ρ of 0.83 [P = 0.015] and 0.86 [P = 0.011], respectively); however, their correlation with baseline PSMA SUVmax and %Δ1 was weaker.
Median gross, clinical, and planning tumor volumes for SBRT were 1.8, 4.6, and 8.8 cm3, respectively. The median of the mean lesion AD delivered by SBRT calculated from voxels within the gross tumor volume) was 28.1 Gy (IQR, 0.5 Gy), whereas the median of the mean BED from SBRT was 116.3 Gy (IQR, 3.8 Gy). The median RE from SBRT was 4.13, reflecting the 9.4 Gy delivered per fraction to the gross tumor volume. Corresponding values for maximum AD delivered by SBRT and maximum BED from SBRT were 29.7 Gy (IQR, 1.0 Gy) and 130.5 Gy (IQR, 6.8 Gy), respectively.
The normalized uptake as measured on planar images from the second cycle was approximately 85%–95% of the uptake measured from the first cycle (normalization with respect to background-corrected uptake in liver, kidneys, and parotid gland; uptake in these organs was assumed to remain unchanged between the 2 cycles). The same lesion regions of interest as drawn on the planar images from the first cycle were used to measure the lesion counts from the second cycle. Lesion shrinkage between the 2 cycles was not accounted for (no significant lesion shrinkage was observed between baseline and interim PSMA PET/CT). If it is assumed that the second cycle delivers 90% of the dose from the first cycle, the median for the mean lesion BED from combined radiopharmaceutical and EBRT (BED from RPT and SBRT) was 159 Gy (IQR, 35 Gy; range, 124–219 Gy), corresponding to a prescription of 33.6 Gy in 3 fractions. The median of the maximum BED from RPT and SBRT was 182 Gy (IQR, 48 Gy; range, 138–258 Gy), corresponding to a prescription of 36.2 Gy in 3 fractions. The total contribution of 2 cycles of RPT to the mean and maximum BED from RPT and SBRT was 34% (range, 13%–52%) and 40% (range, 18%–61%), respectively. If an α/β ratio of 3 is substituted with an α/β ratio of 1.5, the total contribution of [177Lu]Lu-PSMA-617 RPT would decrease to 24% because of its low dose rate. The dosimetry workflow for a representative case is presented in Figures 1–3.
Top: Maximum-intensity projections for 3 PSMA PET/CT scans of 72-y-old patient initially treated with radical retropubic prostatectomy in 1998 (patient 4). Oligorecurrent lymph nodes in lower retrocaval (blue arrow), right posterior common iliac (red arrow), and superior presacral (orange arrow) regions. Bottom: Axial CT slices, centered on lower retrocaval lymph node.
Top: Maximum-intensity projections for 3 [177Lu]Lu-PSMA-617 SPECT/CT scans of patient 4. Oligorecurrent lymph nodes in lower retrocaval (blue arrow), right posterior common iliac (red arrow), and superior presacral (orange arrow) regions. Bottom: Axial CT slices, centered on lower retrocaval lymph node.
Axial views of voxelwise dose maps of patient 4, focused on lower retrocaval lymph node. From left to right: BED (assuming α/β ratio of 3 [BEDα/β = 3]) from first cycle of [177Lu]Lu-PSMA-617 RPT only, from SBRT, and from combined therapy. If second RPT cycle delivers 90% of AD from first cycle, total mean and maximum lesion BEDα/β = 3 would be 219 and 258 Gy, respectively, corresponding to prescription of 40.1 Gy (mean) and 43.9 Gy (maximum) in 3 fractions or EQD2α/β = 3 of 131 Gy (mean) and 155 Gy (maximum). Two cycles of [177Lu]Lu-PSMA-617 RPT were estimated to contribute 52% and 61% to combined mean and maximum BEDα/β = 3, respectively.
Correlations between various indices are presented in Figures 4 and 5. Neither ADRPT nor interim PET SUVmax was significantly correlated with the %Δ2, with a ρ of 0.12 and 0.29, respectively. Analysis was also repeated by substituting SUVmax with SUVpeak (supplemental materials).
Correlations between imaging and dosimetry-based metrics: SPECT SUVmax vs. baseline PSMA PET SUVmax (A), ADRPT vs. baseline PSMA PET SUVmax (B), and ADRPT vs. SPECT SUVmax (C).
Relationship between lesion BED and PSMA PET–assessed response to [177Lu]Lu-PSMA-617 RPT before SBRT (i.e., interim) (A) and after SBRT (i.e., post) (B).
DISCUSSION
The sequential or concurrent combination of cancer therapies may act in an additive and possibly even a synergistic way to increase site-specific coverage of the overall cancer treatment. Focal therapy remains important for omCSPC (3), as the gross sites of disease may be less responsive to complete elimination by systemic therapies. In the present work, a theranostic strategy was implemented in which patients with oligometastatic disease were identified by PSMA PET, which was in turn used to gauge the response and to track the treatment efficacy. This was a feasibility study, a precursor to a phase I trial that will be initiated soon (n = 27 patients; same set of interventions as the current study). The decision to administer 2 cycles of [177Lu]Lu-PSMA-617 RPT was made up front given the increased bioavailability of the radiopharmaceutical in normal tissues due to low-volume metastatic disease in this cohort of patients compared with the VISION trial in which patients had significant castration-resistant disease (8). An effort to be conservative was made given that it was not known how castration-sensitive patients with low-volume disease would tolerate the drug. The increased bioavailability of the radiopharmaceutical due to the limited absorption in small-volume disease needs to be factored in when treating earlier stages of disease. The combination of RPT and SBRT addresses this issue by intensifying the radiation on the target and limiting the exposure of normal tissues. It is likely that the clinical significance of late radiation damage, in particular, radiation nephropathy, is strongly linked to life expectancy. The dosimetry was exploratory and was not used to guide treatment in patients that exhibited lower [177Lu]Lu-PSMA-617 ADs, had a higher PSA doubling rate, or had a poor PSA response to cycles 1 and 2. Treatment intensification could be achieved by increasing the number of cycles (31), increasing the activity per cycle, decreasing the time gap between cycles, or some combination of these. The requirement for accurate dosimetry for optimizing combined-modality approaches is probably most relevant for normal-tissue toxicity. We expect that the importance of dosimetry will increase as RPT is applied earlier in the disease process, especially if the clinical results suggest there is a rationale for dose escalation.
In our previous study with 133 men who received salvage radiotherapy (median of 20 mo after radical prostatectomy), approximately 70% of PSMA PET–positive cases were oligorecurrent (32). The metastasis-selective combination of systemic targeted RPT, followed by focal EBRT, is a novel concept with a strong rationale: both modalities use ionizing radiation and can thus be quantified and compared in terms of BED; RPT with [177Lu]Lu-PSMA-617 enables tumor dose escalation without substantial toxicity to nearby organs; the combination of [177Lu]Lu-PSMA-617 RPT and SBRT allows for the intensification of therapeutic ADs to the target while limiting the exposure of the organs at risk more than if treating with [177Lu]Lu-PSMA-617 alone, escalated to provide similar tumor ADs; [177Lu]Lu-PSMA-617 has the potential to deliver a cytotoxic payload to disseminated occult PSMA-expressing microscopic disease that is below the resolution limits of modern PET scanners; SBRT can provide a high-radiation AD to the larger oligometastatic sites for which the RPT radiation dose might be subtherapeutic; and protracted low-dose radiation might serve as a primer dose that sensitizes cancer cells to subsequent high-dose-rate radiotherapy (33). In the omCSPC setting, the proposed combination therapy may result in the delay of castration resistance and the need for androgen-deprivation therapy. A recent prospective phase III trial concluded that a single-dose radiation therapy (24 Gy, corresponding to a BED [assuming an α/β ratio of 3] of 216 Gy) for the treatment of omCSPC results in a lower 3-y cumulative incidence of local recurrence and distant metastatic progression than does a standard (27 Gy in 3 fractions, corresponding to a BED [assuming an α/β ratio of 3] of 108 Gy) SBRT regimen (∼5% compared with ∼22%, respectively) (34), indicating that comprehensive ablation of oligometastatic lesions, for which [177Lu]Lu-PSMA-617 RPT may help facilitate, is associated with significant mitigation of distant metastatic progression.
A PET-derived ΔSUVmax of at least 75% has been suggested to strongly predict freedom from local failure in patients undergoing oligometastasis-directed ablative radiotherapy (35). The observed median lesion ADRPT of 3.7 Gy/GBq is comparable to previously published results from 10 studies on lesion dosimetry of [177Lu]Lu-PSMA-617 (36). Correlations between aspects of screening PSMA PET and tumor and normal-tissue dose provide a rationale for patient-specific dosing (37). In patients with low-volume metastatic CSPC, the AD in organs appears to be similar or lower in the second [177Lu]Lu-PSMA-617 cycle, suggesting that the bioavailability does not increase in later treatment cycles (10). A trend of decreasing AD in lesions over cycles was previously observed for [177Lu]Lu-PSMA I&T (38).
The 2 main components of the uncertainty in RPT dosimetry are the variability associated with contour delineation and the volume- and sphere-based RC (25,39). A 50% threshold on PET was chosen because it was previously reported to result in the smallest mean differences from morphologic volume measurements (Bland–Altman analysis (40)) and most closely corresponds to volumes delineated on a PSMA PET/MRI scan (41). PET-derived volumes also closely correspond in size to CT-derived volumes; however, in 2 cases, the lesions were not seen on the CT. When lesions were segmented with a 40% or 30% threshold instead of a 50% threshold, the resulting ADRPT was approximately 10% and 40% lower, respectively. Optimal threshold depends on multiple factors such as lesion size and lesion-to-background ratio (42). RCs also depend on factors other than the volume of the object, such as shape and activity distribution, with the dependence being largest for extreme departures from spheric geometry (43). No account was taken of nonuniform RPT dose distribution. Even for large tumors, assessing this issue is limited by the spatial resolution of SPECT, and derived quantities such as dose volume histograms may be more of a reflection of SPECT image limitations than underlying dosimetric nonuniformity. A limitation of the current study is that the inaccuracies associated with quantifying activity in small lesions with SPECT have not been validated via phantom experiments. For objects with sizes comparable to or smaller than the spatial resolution of emission tomographic imaging cameras, substantial underestimation in the apparent radioactivity concentration is observed, leading to, for example, an approximately 50× lower apparent calculated AD for 0.25 cm3 lesions (44). However, the observed good agreement between lesion ADs calculated in Velocity and an additional method using multiple concentric oversized VOIs, which enables activity quantification without partial-volume correction (10,30), increases confidence in the quantification approach as implemented in Velocity.
Although BED does not have a definitive mechanistic underpinning that relates it to cellular biology and does not incorporate reoxygenation, it accounts for the effect of AD rate (14) and is shown to better correlate with toxicity than does AD (45). MIRD pamphlet no. 20 demonstrated equivalent BED response curves for both external-beam data and peptide-receptor data (46). The linear quadratic model is reasonably well validated up to 10 Gy/fraction (47); however, the upper limit of fraction sizes for which the model remains valid is uncertain (48). In ultra-high-dose radiation therapy (>12 Gy fractions), a unique dual-target mechanism of action has been observed that is fundamentally distinct from the classic fractionation model, linking a transient microvascular vasoactive dysfunction to the repression of high-fidelity homologous recombinatory repair of radiation-induced DNA damage (49). Radiosensitivity may also differ within a lesion, between lesions, and among patients and is possibly influenced by prior therapy (50). Moreover, the fractionation response (ratio of radiobiologic parameters α and β, traditionally established from in vitro cell-colony experiments) of prostate cancers has yet to be rigorously defined. To combine the radiobiologic effect of different modalities, the same α/β ratio must be used. Operationally, an α/β ratio of 3 Gy was chosen because it is within the range of reported values for prostate cancer for low-dose-rate brachytherapy and relatively low-dose-per-fraction EBRT (we anticipate that these general findings are likely to be equally applicable to low-dose-rate RPT) and for hypofractionated SBRT (51). Currently, there is no consensus on what radiobiologic parameter values to use for RPT with [177Lu]Lu-PSMA-617 (and if those from EBRT apply) and what is the best approach for relating RPT to EBRT dose response (52). The tumor microenvironment, intratumor genomic heterogeneity, intercellular dose nonuniformity, inflammation- or immune-mediated effects, cell-cycle phase, and chemical factors such as tissue oxygen saturation also impact the biologic response; however, these were not considered in the current work.
CONCLUSION
We demonstrate the feasibility of performing lesional dosimetry in patients with omCSPC undergoing [177Lu]Lu-PSMA-617 RPT followed by SBRT. Combined RPT and SBRT may provide an efficient method to maximize the delivery of meaningful doses to oligometastatic disease while addressing potential microscopic disease reservoirs and limiting the dose exposure of normal tissues.
DISCLOSURE
This study was funded by a Department of Radiology seed grant of Memorial Sloan Kettering Cancer Center with support from Memorial Sloan Kettering Cancer Center’s Radiochemistry and Molecular Imaging Probes Core (NIH/NCI Cancer Center support grant P30 CA008748). The study was also supported by a generous philanthropic gift from the Charles Greenberg Chair Fund. [177Lu]Lu-PSMA-617 was provided as a fully synthesized product from Advanced Accelerator Applications/Endocyte (a division of Novartis). No other potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: Is combination therapy with [177Lu]Lu-PSMA-617 RPT and SBRT for the treatment of omCSPC feasible?
PERTINENT FINDINGS: The proposed combination therapy was well tolerated. Composite lesional dosimetry revealed a relatively high attainable maximum BED of more than 180 Gy, with 2 cycles of [177Lu]Lu-PSMA-617 RPT contributing approximately 40% to the combined maximum BED.
IMPLICATIONS FOR PATIENT CARE: Combined [177Lu]Lu-PSMA-617 RPT and SBRT may provide an efficient method to maximize the delivery of meaningful doses to omCSPC tumors while addressing potential microscopic disease reservoirs and limiting the dose exposure of normal tissues.
Footnotes
Published online Aug. 31, 2023.
- © 2023 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication March 21, 2023.
- Revision received July 11, 2023.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.