Examining the use of the Response Evaluation Criteria in Prostate-specific Membrane Antigen (PSMA) PET/CT (RECIP 1.0) model for assessing the treatment of metastatic castration-resistant prostate cancer, researchers found robust agreement between quantitative application of RECIP via tumor segmentation software and qualitative application of the model through reads by nuclear medicine physicians.
Strong agreement between visual positron emission tomography (PET) scan reads by physicians and quantitative tumor segmentation software in a new study may lead to earlier use in clinical practice of an emerging model for assessing treatment of metastatic castration-resistant prostate cancer (mCRPC).
For the multicenter retrospective study, recently published in Radiology, researchers compared qualitative application (PET scan reads by five nuclear medicine physicians) and quantitative application (qPSMA tumor segmentation software) of the Response Evaluation Criteria in Prostate-specific Membrane Antigen (PSMA) PET/CT (RECIP 1.0) model for assessing the use of lutetium Lu 177 vipivotide tetraxetan (Pluvicto, Novartis) in treating mCRPC.
The patient cohort was comprised of 124 men (median age of 73) with mCRPC, who had a median baseline prostate-specific antigen (PSA) level of 139 ng/mL and had been treated with two cycles of lutetium Lu 177 vipivotide tetraxetan, according to the study. The PET radiotracers gallium 68 PSMA-11 and fluorine 18 rhPSMA-7/7.3 PSMA were administered in 72 percent and 28 percent, respectively, for patients prior to obtaining PET/computed tomography (CT) images at baseline and 12 weeks.
For the differentiation of progressive disease (PD) and non-progressive disease (non-PD) with the RECIP system, the researchers found that agreement between qualitative and quantitative assessments ranged from 92 to 95 percent for the five nuclear medicine readers. The study authors said these results provide a viable alternative to the high interreader variability seen with the four categories of the original RECIP classification system.
“ … In daily practice, only the determination of PD versus non-PD is clinically relevant to identify patients in whom a treatment benefit is not being obtained and thus are candidates for treatment discontinuation,” wrote lead study author Andrei Gafita, M.D., who is affiliated with the Division of Nuclear Medicine and Molecular Imaging at the Russell H. Morgan Department of Radiology and Radiological Science at the Johns Hopkins University School of Medicine in Baltimore, and colleagues.
The study authors also noted that complete interreader agreement in differentiating between PD and non-PD occurred in 83 percent of the cases. While Gafita and colleagues acknowledged a higher rate of interreader agreement with quantitative RECIP (92 percent), they pointed out that current obstacles with tumor segmentation software, such as computational time upward of 6.8 minutes, may delay its implementation in clinical practice.
Additionally, the authors found that in differentiating between PD and non-PD in total tumor volume (TTV), agreement between qualitative and quantitative assessments ranged between 90 to 94 percent for the nuclear medicine readers.
“These findings suggest that the identification of PD in TTV (defined as a 20% increase in TTV relative to baseline) at qualitative reads can be used in practice as a surrogate for quantitative changes derived from tumor segmentation software,” maintained Gafita and colleagues.
Beyond the inherent limitations of a retrospective study, the researchers acknowledged a possible overestimation of the value of the RECIP classification system due to using the same patient cohort for assessment of visual RECIP and quantitative RECIP that was originally used to develop the classification system.