Read the Full Video Transcript

Chris Wallis: Hello, and thank you for joining us for our UroToday journal club. Today, we're discussing an important and recent publication in European Urology entitled, Is Prostate Cancer-Specific Membrane Antigen Positron Emission Tomography/Computed Tomography Imaging Cost-Effective in Prostate Cancer: An Analysis Informed by the proPSMA Trial.

I'm Chris Wallis, a fellow in urologic oncology at Vanderbilt. With me today is Zach Klaassen, Assistant Professor in the Division of Urology, the Medical College of Georgia.
You can see here, the citation for this recent publication in European Urology. When we get into the background for this study, it's clear to most clinicians treating prostate cancer patients that imaging is critical along the whole continuum of prostate cancer care from initial diagnosis, at which time we perform staging before undertaking local therapies, to the detection of recurrence and monitoring response to systemic therapy. Conventional imaging, according to most guideline bodies, involves a CT scan as well as a bone scan. However, this imaging performs generally quite poorly, particularly lacking sensitivity when PSA's are <2 ng/mL.
More recently, novel PET CT radiotracers may address this issue, and data to this effect comes from most prominently, the proPSMA trial led out of Australia. This trial enrolled men with untreated biopsy-proven prostate cancer who were being considered for curative-intent treatment. They had to have a PSA of 20 ng/mL or Gleason Grade Group 3 to 5 or clinical stage T3 in order to increase the risk profile of the cohort and increased event rates.

When we look at the population in the study, we have 302 randomized patients, of which 152 went on the conventional imaging approach and 150 were randomized to 68-gallium-PSMA PET CT. The results of these imaging studies were compared to a reference standard comprising both hard and soft criteria as you can see here. When we look at the primary outcome, PSMA PET CT had 27% greater accuracy than conventional imaging. Not only did it have greater accuracy, when we look at each of the components of interest, looking at the area under the curve or looking at specificity or sensitivity, we see that PSMA PET CT outperformed conventional imaging in detecting any metastatic disease, in detecting pelvic nodal disease and in detecting distant metastases. Further, there was lower calls for equivocal lesions on the basis of PSMA PET CT.
However, before we adopt this into routine clinical practice, it's important to understand not just whether there's a potential for clinical benefit, but whether these benefits are justified by differences in resource utilization. So to do this, the authors undertook a cost-effectiveness analysis, and they based it on the proPSMA trial. So just to reiterate, this was a study that included men with high-risk prostate cancer who are considering a treatment with prostatectomy or radiotherapy. The comparison was conventional imaging using CT and bone scan to 68-gallium-PSMA PET CT. The key primary outcome was diagnostic accuracy.
So for the present study looking at the cost-effectiveness of this approach, the authors undertook a micro-costing analysis. This here is the decision tree comparing PSMA PET CT and conventional imaging. They looked at costs for both production and delivery. For the PSMA PET CT group, they used cost provided by one of the sites in the proPSMA trial, and for conventional imaging, they used the Australian Medicare benefits schedule. The cost and duration of administration were derived from the trial. The authors undertook a societal perspective, which is notable, as most cost-effectiveness analyses only consider the payer perspective.
The outcomes were expressed as accurate diagnoses, and they quantified this as the incremental difference in the probability-adjusted true positives minus false negatives. Probabilities and measures of accuracy were derived from the proPSMA study. The authors then performed a number of sensitivity analyses to test for the effect of variability in radiopharmaceutical generation price, wages for time inputs, and the number of scans performed per radiopharmaceutical elution. At this point in time, I'll hand it over to Zach to take us through the results of this interesting analysis.

Zach Klaassen: Thanks, Chris. This is the summary table from the original proPSMA trial, and it notes the key results that informed this economic evaluation. So on the left, we can see sensitivities and specificities for distant and nodal metastases. We see the conventional imaging on the right and in the middle the gallium-PSMA PET CT arm. Going through this in sequence, the distant metastases sensitivity was 92% for gallium-PSMA and 54% for conventional imaging. Specificity was also high for both, 99% for PSMA and 93 for conventional imaging. When we look at nodal metastases, the sensitivity was 83% for PSMA and only 23 for conventional imaging with high specificity in both arms, 99% for PSMA and 96% for conventional imaging. What's important, and the authors note that this is in terms of patient convenience. Looking at the time spent in terms of the scan duration, only one-and-a-half hours for PSMA and upwards of five-and-a-half hours for conventional imaging.

This table looks at the gallium-PSMA costing that they used for the economic analysis. At the top is consumables in terms of product synthesis, in the middle and subsequently down is the consumables for quality control. You can see that the kit costs 120 Aussie Dollars, $23 for the precursor, 75 for the reagents and chemicals, including the sodium chloride, acetonitrile, PSMA-11, peptide ethanol, distilled water. The accessories, which incorporated the syringes, the hairnets, the gloves, the pH strips, and the filters cost $29. Fixed assets, including the generator lease, the gamma spectrometer, TLC, scanner, gamma counter, et cetera, that cost $603. So by far the highest cost. Staff, $62 for a subtotal in this group of $914. In terms of consumables, you can see here once again, the breakdown with chemicals at $6, accessories at 43, fixed assets at 58, and staff at 118 for a subtotal of 225 and a total of 1,139 Aussie Dollars for this cost analysis.

So looking at the results, the cost per scan, just straight-up cost, $1,140 for the PSMA PET scan as we just discussed, whereas conventional imaging cost $1,181. If you adopted a societal perspective, including delivery time, the cost per scan favored PSMA PET at $1,203 compared to $1,412 for conventional imaging.

When talking about the accuracy of detection for each metastasis type separately, when looking at PSMA PET CT associated with a $959 saved per additional accurate detection of nodal disease versus conventional imaging, which was $1,412 saved per additional accurate detection of distant metastasis compared to conventional imaging. So it's important, this kind of highlights that when we look at the societal perspective, looking at metastasis detection type, this is important for showing the benefit and the cost savings for PSMA PET versus conventional imaging.

This is a tornado plot of deterministic sensitivity analyses. You can see on the left, number of patients per elution, a one to four, cost of conventional imaging, time spent in the gallium-PSMA and conventional imaging group, staff time per synthesis, generator cost, the sensitivity of conventional imaging, productivity cost per hour, sensitivity of conventional imaging and specificity of conventional imaging from a nodal distant metastasis perspective.

Looking at this figure, the black bars, which you can see here, show variation in cost per additional accurate case detected using the minimum range. In looking at the bar from -1,250 to +1,150, in terms of spend versus cost savings, variation towards the left favors the PSMA PET CT imaging modality. In summary for this figure, assuming one patient per gallium-PSMA elution, this resulted in an incremental cost per additional case detected for gallium-PSMA compared with conventional imaging of $2,553 for nodal disease and 3,758 for distant metastasis. This is a weighted cost per accurate diagnosis of $1,138. If we are able to have three patient doses per elution, this would generate $950 saved for additional accurate diagnosis. You can see by just looking at this tornado plot, that one of the main factors in cost is how many patients per elution can be used for this imaging modality.

So several discussion points from this cost-effectiveness analysis of the proPSMA trial. This is the first economic evaluation to directly use data from a large randomized controlled trial that demonstrated the superior accuracy of PSMA PET CT in this setting. This analysis indicates that PSMA PET CT is dominant to conventional imaging in the short term for staging men with high-risk prostate cancer, as it has greater accuracy and is also cheaper. A more accurate diagnosis of metastatic disease with PSMA PET is likely to result in more men receiving appropriate treatment for metastasis and thus not incurring the cost, poor survival, and poor quality of life associated with an incorrect diagnosis.

Several points that are unique to this cost-effective analysis is that the key inputs for the analysis were defined prospectively at the commencement of the proPSMA trial. It also included aspects of producing PSMA radiotracers, which was previously cited as jurisdictional differences in costs, and as you saw in the previous tornado plot, was significant in assessing the cost-effectiveness for PSMA PET versus conventional imaging.

So in conclusion, PSMA PET CT is dominant, with both lower direct comparative costs and greater accuracy when compared with conventional imaging for the detection of metastatic disease in men with high-risk prostate cancer. Combined with the other findings from proPSMA for patient management change, lower radiation exposure, and few equivocal findings, a compelling case can be made for adopting PSMA PET CT in this disease space.

Thank you very much. We hope you enjoyed this journal club today discussing the cost-effectiveness of proPSMA trial. Thank you.