PYLCLARI Solution for injection Ref.[51072] Active ingredients: Piflufolastat ¹⁸F

Pharmacotherapeutic group: Diagnostic radiopharmaceuticals, other diagnostic radiopharmaceuticals for tumour detection
ATC code: V09IX16

Mechanism of action

Prostate-Specific Membrane Antigen (PSMA), is a trans-membrane glycoprotein primarily expressed in normal human prostate epithelium at low levels, but may be overexpressed by malignant tissues, particularly by prostate cancer cells, including metastatic disease. Fluorine (¹⁸F) is a β+ emitting radionuclide that enables positron emission tomography. Piflufolastat (¹⁸F) is a selective secondgeneration fluorine-18-labeled small-molecule PSMA inhibitor. Based on the intensity of the signals, PET images obtained using piflufolastat (¹⁸F) indicate the presence of PSMA expressing tissues.

Pharmacodynamic effects

At the chemical concentrations used for diagnostic examinations, this medicinal product does not appear to have any pharmacodynamic activity.

Clinical efficacy

The safety and efficacy of piflufolastat (¹⁸F) were evaluated in three prospective, open-label, multicenter clinical studies in men with prostate cancer: OSPREY (NCT02981368), CONDOR (NCT03739684), and PYTHON (EudraCT number 2020-000121-37).

OSPREY cohort A enrolled a cohort of 268 men with high-risk biopsy-proven prostate cancer who were considered candidates for radical prostatectomy and pelvic lymph node dissection. Each patient received a single piflufolastat (¹⁸F) PET/CT from mid-thigh to skull vertex. Three central independent readers blinded to all clinical information interpreted each PET scan for the presence of abnormal uptake in pelvic lymph nodes in multiple subregions, including the common iliac lymph nodes. Coprimary endpoints were specificity and sensitivity of piflufolastat (¹⁸F) PET/CT against histopathology within the pelvic lymph nodes. Secondary endpoints were Positive Predictive Value (PPV) and Negative Predictive Value (NPV) of piflufolastat (¹⁸F) PET/CT to predict the presence or absence respectively of prostate cancer within the prostate gland and lymph nodes in Cohort A.

A total of 252 patients (94%) underwent prostatectomy and pelvic lymph node dissection and had sufficient histopathology data for evaluation of the pelvic lymph nodes. Surgical specimens were separated into three regions: left hemipelvis, right hemipelvis, and other. For each patient, piflufolastat (¹⁸F) PET/CT results and histopathology results obtained from dissected pelvic lymph nodes were compared by surgical region. PET/CT results in locations that were not dissected were excluded from analysis. For the 252 evaluable patients, the mean age was 64 years (range 46 to 84 years). The median serum PSA was 9.3 ng/mL. The total Gleason score was 7 for 19%, 8 for 46%, and 9 for 34% of the patients, with the remainder of the patients having Gleason scores of 6 or 10.

The pre-defined thresholds for the co-primary endpoints were 40% for sensitivity and 80% for specificity. Sensitivity did not reach statistical significance for at least 2 of the 3 independent imaging reviewers, therefore, it was considered a failed study.

Table 2 shows piflufolastat (¹⁸F) PET/CT performance by reader using pelvic lymph node histopathology as standard of truth, at the patient-level with region matching (one true positive region defines a true positive patient). Approximately 24% of the evaluable patients had pelvic lymph node metastases based on histopathology (95% confidence interval: 19%, 29%).

Table 2. Performance evaluation of piflufolastat (¹⁸F) PET/CT for pelvic lymph node metastasis detection in OSPREY cohort A (n=252) using Patient-Level and Region-Matched analysis:

Abbreviations: CI = confidence interval, PPV = positive predictive value, NPV = negative predictive value

For primary staging (OSPREY Cohort A), high level reader agreement for pelvic lymph nodes metastases (92.5%) was achieved with Fleiss' kappa statistic of 0.78 (95%CI: 0.71; 0.85).

In exploratory analyses, there were numerical trends towards more true positive results among patients with total Gleason score of 8 or higher and among patients with tumor stage of T2c or higher relative to those patients with lower Gleason score or tumor stage.

A comparison on diagnostic performance of piflufolastat (¹⁸F) PET/CT with baseline conventional imaging (CI) in patients with high risk prostate cancer from Osprey Cohort A was performed as a posthoc study. Piflufolastat (¹⁸F) PET/CT demonstrated a 3-fold higher PPV than conventional imaging (median 86.7% vs. 28.3%, respectively) despite similar sensitivity (median 40.3% for piflufolastat (¹⁸F) PET/CT and 42.6% for conventional imaging). Mean specificiy of piflufolastat (¹⁸F) PET/CT was 97.9% and 65.1% for CI and mean NPV 83.2% vs. 78.8% respectively.

CONDOR enrolled 208 patients with biochemical evidence of suspected recurrent prostate cancer after initial treatment (radical prostatectomy in 85% of the patients). The median serum PSA was 0.82 ng/mL. All enrolled patients had negative or equivocal for prostate cancer conventional imaging evaluation (for most patients, CT or MRI) within 60 days prior to receiving piflufolastat (¹⁸F). All patients received a single PET/CT from mid-thigh to skull vertex with optional imaging of the lower extremities. Three independent central readers, blinded to all clinical information, evaluated each PET/CT scan for the presence and location of positive lesions. Location of each lesion was categorized into 5 regions (prostate/prostate bed, pelvic lymph nodes, other lymph nodes, soft tissue, bone). The primary endpoint was the correct localisation rate (CLR) at the patient level, defined as the percentage of patients for whom there was a one-to-one correspondence between localisation of at least one lesion identified on piflufolastat (¹⁸F) PET/CT imaging and the composite truth standard. If the lower bound of the 95% CI was >0.2 (CLR of 20%) for at least 2 of the 3 independent imaging reviewers, then the primary endpoint analysis was considered a success. The secondary endpoint was the impact on patient management (IMP) defined as the percentage of patients with a change in intended prostate cancer treatment plans due to piflufolastat (¹⁸F) PET/CT as measured by comparison of intended management questionnaires completed pre- and post- piflufolastat (¹⁸F) PET/CT imaging results.

Depending on the reader, a total of 123 to 137 patients (59% to 66%) had at least one lesion that was identified as piflufolastat (¹⁸F) PET-positive (Table 3). The region most commonly observed to have a PET-positive finding was pelvic lymph nodes (40% to 42% of all PET-positive regions) and the least common region was soft tissue (6% to 7%).

Depending on the reader, 99 to 104 patients with a piflufolastat (¹⁸F) PET-positive region had location-matched composite reference standard information that consisted of histopathology, imaging (CT, MRI, ultrasound, fluciclovine (¹⁸F) PET, choline PET, or bone scan) obtained within 60 days of the PET/CT scan, or response of serum PSA level to targeted radiotherapy. Table 3 shows patientlevel performance results of piflufolastat (¹⁸F) PET/CT by reader, including location-matched positive predictive value, also known as Correct Localization Rate (CLR). A patient was considered true positive if they had at least one matching location positive on both piflufolastat (¹⁸F) PET/CT and the composite reference standard.

Table 3. Patient-Level Performance of piflufolastat (¹⁸F) PET/CT in CONDOR (n=208):

Abbreviations: CLR = location-matched positive predictive value, CI = confidence interval

Table 4 shows patient-level piflufolastat (¹⁸F) PET/CT results from the majority read stratified by serum PSA level. Percent PET positivity was calculated as the proportion of patients with a positive PET/CT out of all patients scanned. The likelihood of a patient having at least one piflufolastat (¹⁸F) PET-positive lesion generally increased with higher serum PSA level.

Table 4. Patient-Level piflufolastat (¹⁸F) PET results and percent PET positivity* stratified by serum PSA level in the CONDOR study using majority result among three readers (n=199)**:

^* Percent PET positivity = PET positive patients/total patients scanned. PET positive patients include true positive and false positive patients as well as those who did not have reference standard information.
^** Six patients were excluded from this table due to lack of baseline PSA level, and three patients were excluded from this table due to lack of majority result among three readers.
Abbreviations: TP = true positive, FP = false positive, CI = confidence inte

For the 207 patients with medical management questionnaires completed by treating physicians at preand post-PSMA imaging, 64% (131/207) of patients had a change in intended management after piflufolastat (¹⁸F) PET/CT. Of the patients with changed clinical plans, 79% (103/131) were due to positive PSMA PET/CT findings, and 21% (28/131) were due to negative scans. The most frequent changes were from salvage local therapy to systemic therapy (58 patients), from observation to initiating any therapy (49 patients), from noncurative systemic therapy to salvage local therapy (43 patients), and from planned treatment to observation (no treatment) (9 patients).

PYTHON was a randomised, open-label, two-treatment cross-over study. It enrolled 217 male patients with first biochemical recurrence of prostate cancer, who underwent definitive therapy (radical prostatectomy (RP) ± extented lymph node dissection (eLND) in 73.2% patients, EBRT or brachytherapy in 26.8% patients). The primary endpoint was detection rate (DR) defined as number of patients defined as positive at patient level by the independent readers among the total number of patients assessed (for piflufolastat (¹⁸F) PET/CT and fluorocholine (¹⁸F) PET/CT). A significant difference of 12% detection rate in favour of piflufolastat (¹⁸F) against Fluorocholine (¹⁸F) was predefined. Secondary endpoints were sensitivity (ratio between the number of patients defined as positive for a given region by the independent readers and the total number of patients assessed as positive for a given region by the truth panel), concordance (ratio between the number of regions defined as positive by both piflufolastat (¹⁸F) PET/CT and Fluorocholine (¹⁸F) PET/CT + the number of regions defined as negative by both piflufolastat (¹⁸F) PET/CT and Fluorocholine (¹⁸F) PET/CT and the total number of assessed regions) and impact on patient management.

Two-hundred one patients performed one piflufolastat (¹⁸F) PET/CT and one fluorocholine (¹⁸F) PET/CT from mid-thigh to skull vertex in a randomised order. Three independent central readers, blinded to all clinical information, evaluated each piflufolastat (¹⁸F) and each fluorocholine (¹⁸F) PET/CT for the presence and location of positive lesions. Location of each lesion was categorized into 5 regions (prostate/prostate bed, pelvic lymph nodes, other lymph nodes, bone, soft tissue). Recurrence was detected by the blind read experts in 119 (60.4%) and 82 (41.0%) of the patients with piflufolastat (¹⁸F) and fluorocholine (¹⁸F) PET/CT, respectively. Details of overall independent reader’s interpretation by PSA level is given in Table 5.

Table 5. Per-patient detection rate of PET/CT by PSA level in PYTHON study (N=201):

Per-patient sensitivity was assessed for 37 patients with a standard of truth and is reported in Table 6. Per-patient sensitivity of (¹⁸F)-piflufolastat was significantly higher than that of (¹⁸F)-fluorocholine (p<0.0001).

Table 6. Per-patient sensitivity (n=37):

The concordance rate between (¹⁸F)-piflufolastat PET/CT and (¹⁸F)-fluorocholine PET/CT according to central blind readers, per-region was remarkably high for all regions of interest, namely prostate bed: 87.3% (81.9; 91.3), pelvic lymph nodes: 73.9% (67.3; 79.5), extrapelvic lymph nodes: 86.5% (81.0; 90.6), bones: 86.9% (81.5;91.0), and other organs: 92.0% (87.3; 95.1).

Regarding the localization of recurrence, the central readers achieved an agreement of 84.2% with a Fleiss' kappa statistic of 0.58 (95% CI: 0.47; 0.70) for all biopsy images in OSPREY Cohort B. In CONDOR, the central readers exhibited 76% agreement in interpreting positive or negative piflufolastat (¹⁸F) PET/CT scans with a Fleiss' kappa statistic of 0.65 (95% CI: 0.58; 0.73), while the concordance between each central reader and the local reader ranged from 83% to 84%. In PYTHON, the inter-reader agreement percentage was 67.8%, and the corresponding Fleiss' kappa was 0.55 (95% CI: 0.47; 0.63).

Paediatric population

The European Medicines Agency has waived the obligation to submit the results of studies with Pylclari in all subsets of the paediatric population in diagnosis of prostate cancer (see section 4.2 for information on paediatric use).