Abstract
PURPOSE
There is limited data available on the real world patterns of molecular testing in men with advanced prostate cancer. We thus sought to evaluate Next Generation Sequencing (NGS) testing in the United States, focused on single versus serial NGS testing, the different disease states of the testing (hormone-sensitive vs castration-resistant, metastatic vs non-metastatic), tissue vs plasma circulating tumor DNA assays, and how often actionable data was found on each NGS test.
METHODS
The Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE) clinical-genomic database was used for this retrospective analysis, including 1,597 patients across fifteen institutions. Actionable NGS data was defined as including somatic alterations in homologous recombination repair genes, mismatch repair deficiency, microsatellite instability, or a high tumor mutational burden ≥ 10 mut/MB.
RESULTS
Serial NGS testing (2 or more NGS tests with specimens collected more than 60 days apart) was performed in 9% (n=144) of patients with a median of 182 days in between test results. For the second NGS test and beyond, 82.1% (225/274) tests were from circulating tumor DNA assays and 76.1% (217/285) were collected in the metastatic castration-resistant setting. New actionable data was found on 11.1% (16/144) of second NGS tests, with 3.5% (5/144) tests detecting a new BRCA2 alteration or microsatellite instability. A targeted therapy (PARP inhibitor or immunotherapy) was given after an actionable result on the second NGS test in 31.3% (5/16) of patients.
CONCLUSION
Repeat somatic NGS testing in men with prostate cancer is infrequently performed in practice and can identify new actionable alterations not present with initial testing, suggesting the utility of repeat molecular profiling with tissue or blood of men with mCRPC to guide therapy choices.
Introduction
In men with advanced prostate cancer, international guidelines1,2 now recommend both germline and somatic genetic testing to inform treatment decisions and family counseling given the high prevalence of germline DNA repair mutations in this setting and the availability of FDA approved targeted therapies for actionable somatic alterations.
Mutations in homologous recombination repair genes, particularly BRCA1 and BRCA2, can be targeted with Poly (ADP-Ribose) Polymerase (PARP) inhibitors, such as olaparib3–6, rucaparib7–9, talazoparib10, and niraparib11. Immunotherapy with the PD-1 checkpoint inhibitor pembrolizumab is also FDA approved12,13 and improved outcomes are seen in patients with mismatch repair deficiency14, microsatellite instability15, a high tumor mutational burden16, and possibly CDK12 alterations.17–19
Somatic alterations such as these can be found through tumor testing, which is comprised of a targeted panel of genes that are sequenced using a biopsy or plasma circulating tumor DNA. Next-generation sequencing (NGS) is a technology which enables massively parallel sequencing of DNA or RNA to detect variants/mutations as part of a precision medicine approach. When tissue biopsies are sent for NGS testing, sequencing is successful in 50–60% of cases, depending on the age and type of the sample.20,21 Unsuccessful sequencing is usually due to insufficient tissue and low DNA content21, and this occurs more often with bone biopsies since they require decalcification.22,23
In addition, there are FDA-approved commercial blood-based liquid biopsies, which can detect somatic alterations in cell-free circulating tumor DNA (ctDNA).24–26 While blood-based NGS testing is non-invasive and has a quicker turn-around time27, limitations include false negatives from low tumor content and confounding due to clonal hematopoiesis of indeterminate potential or germline variants.26
In prostate cancer, archival tissue from an initial prostatectomy or a diagnostic biopsy specimen may not necessarily reflect the genotype once a patient has metastatic and/or castration-resistant disease. For instance, AR alterations are rarely seen in therapy-naive localized prostate cancer, but are much more commonly seen in advanced prostate cancer.28–30 Other mutations, including DNA damage repair alterations, appear to be early somatic events or are present in the germline.15,30,31 Tumor evolution in prostate cancer may also be shaped by systemic therapies, such as androgen deprivation therapy (ADT), docetaxel chemotherapy, and androgen receptor pathway inhibitors.32,33
We performed a multi-center retrospective analysis of 1,597 patients in the Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE)34 database in order to characterize patterns of single versus serial NGS testing. We looked at the timing and type of NGS tests ordered at different points in a patient’s disease course, and calculated how often actionable mutations were found.
Methods
Patient population
A retrospective analysis was conducted using the Prostate Cancer Precision Medicine Multi-Institutional Collaborative Effort (PROMISE) database, which includes deidentified clinical and genomic data. Patients’ race was self-identified. Patients had genomic testing (tissue or blood cell-free circulating tumor DNA) through CLIA certified commercially available platforms during routine clinical care. Data were collected from registered institutions between 4/1/2020 and 1/25/2023 using a standardized RedCap database. This study was approved by local institutional review boards at participating sites per institutional policy and the Declaration of Helsinki.
At the time of analysis, there were 1,912 patients with prostate cancer treated at fifteen academic centers and 315 patients were excluded (Fig 1). In total, there were 1,597 patients that were included (Supplemental Table 1). In addition to excluding germline and AR-V7 tests, we excluded NGS entries which were empty (site data entrant may have been in the process of entering NGS data or erroneously created an NGS page) or incomplete (unknown test result dates, unknown specimen source). We also filtered out NGS tests where reports indicated insufficient quality for testing (including key words such as “failed,” “reduced,” “quality,” “insufficient,” and “incomplete”). Each NGS test report fulfilling this instance was manually reviewed before the patient was excluded.
Figure 1.
Patient grouping. n indicates the number of patients and t indicates the number of Next Generation Sequencing (NGS) tests. Included patients were grouped by two branch points, with the first being whether they received 1 NGS test or 2 or more NGS tests. If a patient had 1 NGS test, they were sub-grouped into either single tissue or single blood (plasma including cell free DNA and circulating tumor DNA assays). If a patient had 2 or more NGS tests, the days between the specimen collection dates were compared. If the specimen collection dates were within 60 days of each other, this was considered concurrent NGS testing. If more than 60 days passed before a second specimen was collected, the patient was placed in the serial testing group.
NGS-associated patient characteristics
The PROMISE database includes annotation of a patient’s disease state (hormone-sensitive vs castration-resistant, metastatic vs non-metastatic) at the time each sample utilized for NGS testing was obtained. We categorized the disease states as non-metastatic hormone-sensitive prostate cancer (nmHSPC), metastatic hormone-sensitive prostate cancer (mHSPC), non-metastatic castration-resistant prostate cancer (nmCRPC), or metastatic castration-resistant prostate cancer (mCRPC). Serial NGS testing was defined as when a patient received two or more NGS tests with specimen collection separated by 60 or more days, whereas concurrent testing was when the NGS specimen collection dates overlapped by 60 or less days.
Statistical methods
For the primary objective, we hypothesized that new actionable alterations would be found with serial NGS testing, including both tissue and plasma assays. Baseline demographics were summarized by median for continuous variables and by frequencies and relative frequencies for categorical variables. The relationship between different NGS tests were assessed using the Chi-squared test or Fisher’s exact test for categorical variables and the Wilcoxon test was used for continuous variables. Differences were considered significant at confidence intervals greater than 95% (p < 0.05).
Actionable data
Actionable data was defined as including somatic alterations in homologous recombination repair (HRR) genes that align with the FDA indications for PARP inhibitors (BRCA1, BRCA2, ATM, BRIP1, BARD1, CDK12, CHEK1, CHEK2, FANCL, PALB2, PPP2R2A, RAD51B, RAD51C, RAD51D, or RAD54L) or mismatch repair (MMR) genes (MSH2, MSH3, MSH6, MLH1, MLH3, PMS1, or PMS2), loss of the listed MMR genes protein expression on immunohistochemistry, microsatellite instability (MSI-high), or a high tumor mutational burden (TMB) defined as ≥ 10 mut/MB.
Results
Patients in the single versus serial NGS testing groups
Of the 1,597 patients analyzed in the PROMISE database, most patients received a single NGS test (n=1398; 87.5%, Figure 1). Single tissue-based testing was completed in 59.9% (957/1398) patients and blood-based circulating tumor (ct) DNA testing in 27.6% (441/1398). For the single tissue group, 50.1% (484/957) of specimens were from the prostate (Supplemental Table 1). For all patients, 67.9% (1084/1597) of patients had at least one tissue test.
Among patients who received multiple NGS tests, the serial group contained 9% of patients (144/1597), defined as two or more NGS tests with specimens collected more than 60 days apart. The first test in the serial group was 63.9% (92/144) ctDNA specimens, 22.2% (32/144) from the prostate, and 13.9% (20/144) were from metastatic tissue specimens (Supplemental Table 1). The second test in the serial group was 70.8% (102/144) ctDNA specimens, 16.0% (23/144) from the prostate, and 13.2% (19/144) were from metastatic tissue specimens. The serial group contained 72.2% (104/144) Caucasian patients and 19.4% (28/144) Black patients. The mean number of serial NGS tests by race was not significantly different (Caucasian: 3.95 (95% CI 3.43–4.47) and Black: 4.04 (95% CI 3.36–4.72); p=0.83).
Baseline characteristics were evaluated between the single test and serial NGS test groups (Table 1). T stage at diagnosis differed (p=<0.0001), likely due to the difference in Tx (50.7% in the serial group and 30.6% in the single NGS test group). The mean number of systemic treatments after the first NGS test was higher in the serial NGS test group at 1.65 compared to 1.40 in the single test group (p=0.012). Treatment with PARP inhibitors or immunotherapy did not differ, but both taxane and platinum chemotherapy was given more frequently in the serial NGS group (p=<0.0001 and p=0.0003, respectively). Median follow-up time after the first NGS test result date was not significantly different (serial NGS group: 35.8 months (95% CI 22.1–40.8) vs single NGS group: 24.9 months (95% 21.6–31.2); p=0.80).
Table 1.
Patient Characteristics
| Single test n = 1398 (%) | Serial tests n = 144 (%) | p-value | ||
|---|---|---|---|---|
| Race n (%) | Caucasian | 993 (71.0%) | 104 (72.2%) | 0.93 |
| Black | 276 (19.7%) | 28 (19.4%) | ||
| Other | 129 (9.2%) | 12 (8.3%) | ||
| Median age at diagnosis (interquartile range) | 63 (57–69) | 61 (54–67) | 0.03 | |
| Gleason at diagnosis n (%) | 6–7 | 359 (25.7%) | 39 (27.1%) | 0.90 |
| 8–10 | 763 (54.6%) | 79 (54.9%) | ||
| Missing | 276 (19.7%) | 26 (18.1%) | ||
| Median PSA at diagnosis (interquartile range) | 18.25 (7.18–94.75) | 25.8 (8.7–146.95) | 0.09 | |
| Histology n (%) | Adeno | 1241 (88.8%) | 129 (89.6%) | 0.26 |
| Neuro | 10 (0.72%) | 2 (1.4%) | ||
| Mixed | 2 (0.14%) | 1 (0.69%) | ||
| Missing | 145 (10.4%) | 12 (8.3%) | ||
| T stage at diagnosis n (%) | T1–T2 | 484 (34.6%) | 32 (22.2%) | <0.0001 |
| T3–T4 | 322 (23.0%) | 28 (19.4%) | ||
| Tx | 428 (30.6%) | 73 (50.7%) | ||
| Missing | 164 (11.7%) | 11 (7.6%) | ||
| N stage at diagnosis n (%) | N0 | 445 (31.8%) | 40 (27.8%) | 0.28 |
| N1 | 348 (24.9%) | 45 (31.3%) | ||
| Nx | 440 (31.5%) | 47 (32.6%) | ||
| Missing | 165 (11.8%) | 12 (8.3%) | ||
| M stage at diagnosis n (%) | M0 | 469 (33.6%) | 62 (43.1%) | 0.10 |
| M1 | 429 (30.7%) | 43 (29.9%) | ||
| Mx | 336 (24.0%) | 27 (18.8%) | ||
| Missing | 164 (11.7%) | 12 (8.3%) | ||
| De novo metastatic n (%) | Yes | 588 (42.1%) | 59 (41.0%) | 0.59 |
| No | 681 (48.7%) | 76 (52.8%) | ||
| Missing | 129 (9.2%) | 9 (6.3%) | ||
| Sites of first metastases n (%) | Pelvic LN | 553 (39.6%) | 66 (45.8%) | 0.15 |
| Distant LN | 431 (30.8%) | 43 (29.9%) | 0.85 | |
| Bone | 934 (66.8%) | 94 (65.3%) | 0.71 | |
| Liver | 78 (5.6%) | 6 (4.2%) | 0.57 | |
| Lung | 104 (7.4%) | 8 (5.6%) | 0.50 | |
| Bladder | 52 (3.7%) | 3 (2.1%) | 0.48 | |
| CNS | 6 (0.43%) | 0 (0%) | ||
| Other | 60 (4.3%) | 6 (4.2%) | 1.00 | |
| Primary definitive therapy n (%) | Radical Prostatectomy | 543 (38.8%) | 56 (38.9%) | 0.93 |
| Definitive Radiation | 609 (43.6%) | 73 (50.7%) | 0.11 | |
| Brachytherapy | 57 (4.1%) | 7 (4.9%) | 0.66 | |
| Cryotherapy | 27 (1.9%) | 0 (0%) | ||
| None | 162 (11.6%) | 8 (5.6%) | ||
| Mean # of systemic treatments after the first NGS test result per patient | 1.40 | 1.65 | 0.012 | |
| Systemic treatments after the first NGS test result | Total | 677 | 178 | |
| PARP inhibitor | 70 | 14 | 0.40 | |
| Immunotherapy | 64 | 13 | 0.46 | |
| Taxane | 479 | 116 | <0.0001 | |
| Platinum | 64 | 35 | 0.0003 |
Serial NGS testing patterns across academic institutions, by specimen type, and by disease state
Serial NGS testing was heterogeneous around the US (Supplemental Table 2). Three institutions (TU, UCSD, UCSF) performed serial testing in >10% of patients. Across all institutions, 82.1% (225/274) of second NGS tests and beyond were from blood ctDNA assays (Figure 2A, Supplemental Table 1). The most common sequence of serial tests was a blood ctDNA assay for both the first test and second tests (BB: 19.4%, 28/144) (Figure 2B). The median interval between serial test results was 182 days (range 119–276) (Supplemental Figure 1).
Figure 2.
(A) Sankey flow diagram showing crossover between blood (plasma circulating tumor DNA assays) and tissue tests with the percentage of tests in bold and percentage with an actionable finding underneath. (B) The most common serial sequencing patterns, where B represents a blood test and T represents a tissue test i.e. TB indicates the first Next Generation Sequencing (NGS) test was tissue and the second NGS test was blood (plasma).
The percentage of first NGS tests in the serial group with an actionable finding was 22.2% (32/144) with 28.8% (15/52) actionable tissue tests and 18.5% (17/92) actionable ctDNA tests (Figure 2A, Table 2). For comparison, 32.8% (314/957) of the single tissue NGS tests and 26.5% (117/441) of the single blood ctDNA tests had an actionable finding (Supplemental Table 3). One difference in the single tissue, single ctDNA, and first test for the serial NGS groups was in the detection of BRCA1/2 (12.0%, 7.7%, and 6.9%, respectively).
Table 2.
Actionable findings on serial NGS testing
| BRCA1/2 | Non-BRCA HRR | MSI-hi/MMRd | TMB-high | Total | |
|---|---|---|---|---|---|
| Patients | n = 144 | ||||
| 1st NGS test |
BRCA2 = 6 (4.2%) BRCA2 and CDK12 = 1 (0.7%) BRCA1 = 2 (1.4%) BRCA1 and ATM = 1 (0.7%) Total = 10 (6.9%) |
CDK12 = 6 (4.2%) ATM = 4 (2.8%) CHEK2 = 1 (0.7%) RAD51D = 1 (0.7%) FANCA = 1 (0.7%) Total = 13 (9.0%) |
MSI-hi = 3 (2.1%) MMRd = 4 (2.8%) Total = 7 (4.9%) |
2 (1.4%) | 32 (22.2%) |
| 2nd NGS test |
BRCA2 = 2 (1.4%) BRCA2 and ATM = 1 (0.7%) Total = 3 (2.1%) |
ATM = 4 (2.8%) CDK12 = 1 (0.7%) ATM and CDK12 = 1 (0.7%) PALB2 = 1 (0.7%) FANCA = 1 (0.7%) Total = 8 (5.6%) |
MSI-hi = 2 (1.4%) | 3 (2.1%) | 16 (11.1%) |
| n = 55 | |||||
| 3rd NGS test |
ATM = 1 (1.8%) CHEK2 = 2 (3.6%) PPP2R2A = 1 (1.8%) Total= 4 (7.3%) |
MSI-hi = 1 (1.8%) MMRd = 1 (1.8%) Total = 2 (3.6%) |
1 (1.8%) | 7 (12.7%) | |
| n = 75 | |||||
| 4th NGS test and beyond |
ATM = 2 (2.7%) FANCA = 1 (1.3%) CHEK2 = 1 (1.3%) Total = 4 (5.3%) |
5 (6.7%) | 9 (12%) | ||
| n = 10 | n = 13 | n = 7 | n = 2 | n = 32 | |
| Received a targeted therapy* after the 1st NGS test |
BRCA2 = 4 (40%) BRCA1 = 1 (10%) Total = 5 (50%) |
CDK12 = 3 (23.1%) (PARPi = 1/ IO = 1 / PARPi & IO = 1) ATM (IO) = 1 (7.7%) RAD51D (PARPi & IO) = 1 (7.7%) Total = 5 (38.5%) |
MSI-high = 3 (42.3%) | 13 (40.6%) | |
| n = 3 | n = 8 | n = 2 | n = 3 | n = 16 | |
| Received a targeted therapy after the 2nd NGS test | BRCA2 = 2 (67%) |
PALB2 = 1 (12.5%) CDK12 (IO) = 1 (12.5%) |
MSI-high = 1 (50%) | 5 (31.3%) | |
targeted therapy = PARP inhibitor (PARPi) or immunotherapy (IO)
In the serial group, 43.8% (63/144) of the first NGS test specimens were collected in the metastatic castration-resistant prostate cancer (mCRPC) setting (Supplemental Figure 2). For the second and third NGS tests, 72.2% (104/144) and 83.7% (45/55) were collected in the mCRPC setting. Overall, 76.1% (217/285) of second NGS tests and beyond were collected in the mCRPC setting. In comparison, for patients who received a single tissue test, more specimens were collected in the nmHSPC setting (32.5%, 311/957) and less in the mCRPC setting (33.0%, 316/957) (Supplemental Table 4). For single blood tests, the vast majority were collected in the mCRPC setting (67.3%, 297/441).
Repeat NGS testing finds new somatic gene mutations in BRCA1/2, non-BRCA HRR, microsatellite instability, mismatch repair deficiency, and high tumor mutational burden
An Oncoprint was generated to summarize the cumulative actionable data detected on repeat NGS testing (includes results from the second test and beyond), which was 30% (43/144 patients) (Figure 3A). Some patients had multiple actionable findings over time, but the most common was in non-BRCA HRR genes at 15% (22/144). New actionable data was found on 11.1% (16/144) of second NGS tests for the serial group (Table 2).
Figure 3.
(A) Oncoprint showing actionable items found on subsequent NGS testing (second NGS and beyond) and treatments given after the second NGS test result. Actionable is a composite of BRCA1/2 alterations, non-BRCA HRR alterations, microsatellite instability (MSI-high) and mismatch repair deficiency (MMRd), and high tumor mutational burden ≥ 10 mut/MB (TMB). (B) The number of NGS tests that showed a gene alteration (left), each gene’s detection frequency (right), and detection frequency for TMB, MSI-high, and MMRd, color coded by NGS test number.
Importantly, while BRCA2 is found 70% (7/10) of the time on the first NGS test, 30% (3/10) of cases were detected on the second NGS test (Figure 3B, Table 2). BRCA1 was detected 100% (3/3) of the time on the first NGS test and not found on repeat testing. In the non-BRCA HRR genes, ATM, CDK12, FANCA, PALB2, and PP2R2A are found on repeat testing.
Microsatellite instability was newly found 33% (2/6) of the time on the second NGS test (Figure 3B, Table 2). Mismatch repair deficiency was rarely found after the first NGS test with 80% (4/5) detection on the first test. A high tumor mutational burden could be found even past the third NGS test. Some tests (2–4%) had multiple, overlapping actionable items (Supplemental Figure 3).
A swimmer’s plot shows when specimens were collected for NGS testing in relation to disease state and therapies, with actionable NGS test results annotated on the side (Figure 4, Supplemental Figure 4). For patients with an actionable second NGS test, 56.3% (9/16) had ctDNA as the first test (Supplemental Table 5).
Figure 4.
Swimmer’s plot showing specimen collection (black dot) for NGS testing in relation to disease state and therapies (PARP inhibitors, platinum chemotherapy, taxane chemotherapy, and immunotherapy). Pertinent findings on NGS testing are annotated on the side and color-coded by BRCA1/2 alterations (black), non-BRCA homologous recombination repair (HRR) gene alterations (gold), and mismatch repair deficient (MMRd) or microsatellite instability (green).
Patients who received targeted therapy after an actionable finding on the second NGS test
TU10138 had a second Guardant ctDNA test that showed a PALB2 S578T alteration. This was new from their first Guardant ctDNA test which had only shown a TP53 R273S mutation. He was started on olaparib 39 days after the second NGS test result. His PSA was 58.3 ng/mL and nadired at 26.0 while on olaparib. He was on olaparib for a total of 112 days, went on to receive Actinium-225 as the next line of therapy, and died 450 days after starting olaparib.
SF10096’s second NGS test was a Foundation liquid ctDNA test that showed a BRCA2 S2710fs*23 and ATM G2891D mutation. His first NGS test was a Guardant ctDNA test that showed a NF1 splice SNV. He started olaparib 497 days after his second NGS test result, and after enzalutamide and abiraterone. This patient was on olaparib for 305 days (300mg BID for 50 days, 150mg BID for 255 days), with treatment discontinuation due to cytopenias. The patient died 782 days after starting olaparib.
Additional narrative histories are provided in the Supplemental Results for TU10138 and SF10096, and patients who received targeted therapy without benefit: TU10007 (3 cycles of pembrolizumab after the second NGS test showed microsatellite instability; PSA and CT progression), KM10039 (olaparib for 28 days after the second NGS test showed a new BRCA2 A1327fs; PSA progression), and TU10037 (2 cycles of pembrolizumab after the second NGS test showed a CDK12 R93* mutation; PSA and CT progression).
New vs undetected actionable items with serial NGS testing
Actionable items were compared to the preceding test(s) to further classify if a subsequent NGS test detected a new actionable item when none had been detected before (blue) or in addition to a different prior actionable finding (purple) (Figure 5A). NGS tests were also classified as having undetected an actionable item when no item was detected on a subsequent test, but one had been found before (green). For instance, on the second NGS test, 11.1% (16/144) of tests had an actionable item and 8.3% (12/144) tests did not detect the actionable item that had been found on the first NGS test (Figure 5A).
Figure 5.
(A) For each NGS test, the number of patients was sub-divided into four categories based off whether an actionable item was found or not found, and whether the previous NGS test(s) had found an actionable item. (B) The percentage of actionable findings on the second NGS test, when comparing the sequence of the first two tests. B represents a blood (plasma circulating tumor DNA) test and T represents a tissue test i.e. BB indicates the first NGS test was blood ctDNA and the second NGS test was also blood ctDNA.
NGS tests were also compared looking at the first two NGS tests and whether there was crossover between the type of specimen that was used, i.e. a tissue-based test could be followed by a second tissue-based test (T-T) or a blood-based ctDNA test (T-B). The second NGS test in a B-B sequence found new actionable items in 9.6% (7/72) of cases, compared to 10% (5/20) in B-T, 13.4% (4/30) in T-B, and 13.6% (3/22) in T-T (Figure 5B). Undetected actionable items on the second NGS test were not seen in the the T-T sequence.
Discussion
Here we have shown real world practice patterns for tumor genetic testing in men with advanced prostate cancer, describing the disease settings that the tests were performed, the number of tests, and the heterogeneity in serial testing. By examining these practice patterns in the large multicenter PROMISE clinical-genomic database, we provide data on the prevalence of testing and results using tissue and plasma samples across fifteen academic institutions.
Our retrospective real world data demonstrates that serial testing is infrequently done, with only 9% of patients getting more than one somatic NGS test, and with striking heterogeneity between institutions for serial testing practices. As expected, most of the second NGS tests and beyond are ctDNA tests (82.1%) collected in the mCRPC setting (76.1%). When serial testing is done, tests are usually separated by 6–8 months.
Importantly, 11.1% of patients had a new actionable alteration on a second NGS test, which adds upon the 22.2% actionable finding rate from the first NGS test. For BRCA2, we saw that 30% of cases were detected on the second test and microsatellite instability was newly found 33% of the time on the second test. Prior work suggests that the majority (>75%) of MSI-high mCRPC cases are acquired rather than hereditary Lynch Syndrome, and at least 1/3 of cases may be late events in tumor evolution, requiring repeat testing to identify men with mCRPC who may benefit from PD-1 inhibition.15
Many of these new alterations were identified when the first NGS test was plasma ctDNA based rather than tissue, and this is supported by our data on the single tissue and single ctDNA testing groups as well. In the single tissue group, there was a 32.8% actionable finding rate and in the single ctDNA group, this was 26.5%. Low ctDNA abundance and tumor burden reduce the ability to detect actionable alterations, particularly copy losses, while repeat tumor tissue testing may avoid this issue. The detection of BRCA2 and other homologous repair alterations in serial testing permits access to potentially life-prolonging precision therapies such as PARP inhibition alone or in combination with AR inhibition in the mCRPC setting.35–37
A targeted therapy was given after the second NGS test result in almost a third (31%) of patients. Our previous work has shown that Black men with mCRPC are more likely to be found to be MMRd or MSI-high, but receive targeted therapy less frequently than White men.38 Further research into repeat NGS testing in Black men is needed to better understand disparities to access in precision medicine.
Limitations of this study include that NGS testing patterns in academic institutions may not necessarily represent community practice. There may also be a selection bias in terms of access and other socioeconomic factors for the patients in this database. Our real world data contains heterogeneity in terms of whether tissue versus blood assays were used, and also heterogeneity in terms of what testing platform was used that may have varying sensitivity in detecting mutational events. Lastly, since this study was based on real-world electronic health record data, it may contain incomplete or misclassified data which could be a source of potential bias.
In the PROMISE clinical-genomic database, our real world data shows that serial NGS testing in men with prostate cancer is infrequently performed. However, repeat somatic testing can identify new actionable alterations that may have been missed or were not present with initial testing, particularly in BRCA2 or microsatellite instability. Our results suggest the utility of repeat molecular profiling with tissue or blood of men with mCRPC to guide therapy choices.
Supplementary Material
Supplemental Figure 3. A Venn diagram of the overlap between HRD (BRCA1/2 and non-BRCA HRR alterations), MSI-hi/MMRd, and TMB.
Supplemental Figure 4. Swimmer’s plot showing specimen collection (black dot) for NGS testing in relation to disease state and therapies (PARP inhibitors, platinum chemotherapy, taxane chemotherapy, and immunotherapy). Pertinent findings on NGS testing are annotated on the side and color-coded by BRCA1/2 alterations (black), non-BRCA homologous recombination repair (HRR) gene alterations (gold), and mismatch repair deficient (MMRd) or microsatellite instability (MSI-hi; green). Bolded findings indicate those found on the second NGS test. Boxed are patients who received a targeted therapy (PARP inhibitor or immunotherapy) after a second NGS test.
Supplemental Figure 1. The number of days between each serial NGS test result.
Supplemental Figure 2. Sankey flow diagram showing disease state at the time of specimen collection for patients in the serial Next Generation Sequencing (NGS) group. Disease states include metastatic castration-resistant prostate cancer (mCRPC), non-metastatic castration-resistant prostate cancer (nmCRPC), metastatic hormone-sensitive prostate cancer (mHSPC), and non-metastatic hormone-sensitive prostate cancer (nmHSPC).
Context Summary.
Key objective:
We determined how often serial tumor or ctDNA testing results in actionable genetic alterations in men with metastatic prostate cancer. Actionable NGS data was defined as somatic alterations in homologous recombination repair genes, mismatch repair deficiency, microsatellite instability, or a high tumor mutational burden ≥ 10 mut/MB.
Knowledge generated:
In this multicenter retrospective analysis of the PROMISE registry of 1,597 men with metastatic prostate cancer, we identified 9% (n=144) who underwent serial next-generation sequencing (NGS). We identified actionable NGS results in 18–32% of men at initial testing depending on the type of test. Importantly, we found that 11% of serial tests identified new previously undetected and actionable genetic alterations including BRCA2 alterations or microsatellite instability which impact patient management and have major outcome implications.
Relevance:
Serial ctDNA testing in the mCRPC setting can identify new actionable alterations in a significant minority of patients, with 30% of all BRCA2 alterations identified only on repeat testing.
Acknowledgements:
We would like to thank Alyssa Ghose, Brandon Holland, Michael Pierro, Alleda Mack, and Erik Hernandez for their assistance with the PROMISE database, as well the PROMISE investigators and study teams.
Funding Sources:
Andrew J. Armstrong - NCI/NIH R01: 5R01CA233585-05
References
- 1.Schaeffer EM, Srinivas S, Adra N, et al. : NCCN Guidelines(R) Insights: Prostate Cancer, Version 1.2023. J Natl Compr Canc Netw 20:1288–1298, 2022 [DOI] [PubMed] [Google Scholar]
- 2.Parker C, Castro E, Fizazi K, et al. : Prostate cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 31:1119–1134, 2020 [DOI] [PubMed] [Google Scholar]
- 3.de Bono J, Mateo J, Fizazi K, et al. : Olaparib for Metastatic Castration-Resistant Prostate Cancer. N Engl J Med 382:2091–2102, 2020 [DOI] [PubMed] [Google Scholar]
- 4.Hussain M, Mateo J, Fizazi K, et al. : Survival with Olaparib in Metastatic Castration-Resistant Prostate Cancer. N Engl J Med 383:2345–2357, 2020 [DOI] [PubMed] [Google Scholar]
- 5.Mateo J, Porta N, Bianchini D, et al. : Olaparib in patients with metastatic castration-resistant prostate cancer with DNA repair gene aberrations (TOPARP-B): a multicentre, open-label, randomised, phase 2 trial. Lancet Oncol 21:162–174, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mateo J, Carreira S, Sandhu S, et al. : DNA-Repair Defects and Olaparib in Metastatic Prostate Cancer. N Engl J Med 373:1697–708, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Abida W, Campbell D, Patnaik A, et al. : Non-BRCA DNA Damage Repair Gene Alterations and Response to the PARP Inhibitor Rucaparib in Metastatic Castration-Resistant Prostate Cancer: Analysis From the Phase II TRITON2 Study. Clin Cancer Res 26:2487–2496, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Abida W, Patnaik A, Campbell D, et al. : Rucaparib in Men With Metastatic Castration-Resistant Prostate Cancer Harboring a BRCA1 or BRCA2 Gene Alteration. J Clin Oncol 38:3763–3772, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Fizazi K, Piulats JM, Reaume MN, et al. : Rucaparib or Physician’s Choice in Metastatic Prostate Cancer. N Engl J Med 388:719–732, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.de Bono JS, Mehra N, Scagliotti GV, et al. : Talazoparib monotherapy in metastatic castration-resistant prostate cancer with DNA repair alterations (TALAPRO-1): an open-label, phase 2 trial. Lancet Oncol 22:1250–1264, 2021 [DOI] [PubMed] [Google Scholar]
- 11.Smith MR, Scher HI, Sandhu S, et al. : Niraparib in patients with metastatic castration-resistant prostate cancer and DNA repair gene defects (GALAHAD): a multicentre, open-label, phase 2 trial. Lancet Oncol 23:362–373, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hansen AR, Massard C, Ott PA, et al. : Pembrolizumab for advanced prostate adenocarcinoma: findings of the KEYNOTE-028 study. Ann Oncol 29:1807–1813, 2018 [DOI] [PubMed] [Google Scholar]
- 13.Antonarakis ES, Piulats JM, Gross-Goupil M, et al. : Pembrolizumab for Treatment-Refractory Metastatic Castration-Resistant Prostate Cancer: Multicohort, Open-Label Phase II KEYNOTE-199 Study. J Clin Oncol 38:395–405, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Le DT, Durham JN, Smith KN, et al. : Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 357:409–413, 2017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Abida W, Cheng ML, Armenia J, et al. : Analysis of the Prevalence of Microsatellite Instability in Prostate Cancer and Response to Immune Checkpoint Blockade. JAMA Oncol 5:471–478, 2019 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Graf RP, Fisher V, Weberpals J, et al. : Comparative Effectiveness of Immune Checkpoint Inhibitors vs Chemotherapy by Tumor Mutational Burden in Metastatic Castration-Resistant Prostate Cancer. JAMA Netw Open 5:e225394, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Antonarakis ES, Isaacsson Velho P, Fu W, et al. : CDK12-Altered Prostate Cancer: Clinical Features and Therapeutic Outcomes to Standard Systemic Therapies, Poly (ADP-Ribose) Polymerase Inhibitors, and PD-1 Inhibitors. JCO Precis Oncol 4:370–381, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schweizer MT, Ha G, Gulati R, et al. : CDK12-Mutated Prostate Cancer: Clinical Outcomes With Standard Therapies and Immune Checkpoint Blockade. JCO Precis Oncol 4:382–392, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wu YM, Cieslik M, Lonigro RJ, et al. : Inactivation of CDK12 Delineates a Distinct Immunogenic Class of Advanced Prostate Cancer. Cell 173:1770–1782 e14, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hussain M, Corcoran C, Sibilla C, et al. : Tumor Genomic Testing for >4,000 Men with Metastatic Castration-resistant Prostate Cancer in the Phase III Trial PROfound (Olaparib). Clin Cancer Res 28:1518–1530, 2022 [DOI] [PubMed] [Google Scholar]
- 21.Thompson JC, Yee SS, Troxel AB, et al. : Detection of Therapeutically Targetable Driver and Resistance Mutations in Lung Cancer Patients by Next-Generation Sequencing of Cell-Free Circulating Tumor DNA. Clin Cancer Res 22:5772–5782, 2016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sailer V, Schiffman MH, Kossai M, et al. : Bone biopsy protocol for advanced prostate cancer in the era of precision medicine. Cancer 124:1008–1015, 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zheng G, Lin MT, Lokhandwala PM, et al. : Clinical mutational profiling of bone metastases of lung and colon carcinoma and malignant melanoma using next-generation sequencing. Cancer Cytopathol 124:744–753, 2016 [DOI] [PubMed] [Google Scholar]
- 24.Jahr S, Hentze H, Englisch S, et al. : DNA fragments in the blood plasma of cancer patients: quantitations and evidence for their origin from apoptotic and necrotic cells. Cancer Res 61:1659–65, 2001 [PubMed] [Google Scholar]
- 25.Vandekerkhove G, Wyatt A: Circulating tumour DNA as a biomarker source in metastatic prostate cancer. Société Internationale d’Urologie Journal 1:39–48, 2020 [Google Scholar]
- 26.Kwan EM, Wyatt AW, Chi KN: Towards clinical implementation of circulating tumor DNA in metastatic prostate cancer: Opportunities for integration and pitfalls to interpretation. Front Oncol 12:1054497, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lee Y, Clark EW, Milan MSD, et al. : Turnaround Time of Plasma Next-Generation Sequencing in Thoracic Oncology Patients: A Quality Improvement Analysis. JCO Precis Oncol 4, 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Annala M, Vandekerkhove G, Khalaf D, et al. : Circulating Tumor DNA Genomics Correlate with Resistance to Abiraterone and Enzalutamide in Prostate Cancer. Cancer Discov 8:444–457, 2018 [DOI] [PubMed] [Google Scholar]
- 29.Mateo J, Seed G, Bertan C, et al. : Genomics of lethal prostate cancer at diagnosis and castration resistance. J Clin Invest 130:1743–1751, 2020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Robinson D, Van Allen EM, Wu YM, et al. : Integrative clinical genomics of advanced prostate cancer. Cell 161:1215–1228, 2015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Schweizer MT, Sivakumar S, Tukachinsky H, et al. : Concordance of DNA Repair Gene Mutations in Paired Primary Prostate Cancer Samples and Metastatic Tissue or Cell-Free DNA. JAMA Oncol 7:1–5, 2021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Annala M, Taavitsainen S, Khalaf DJ, et al. : Evolution of Castration-Resistant Prostate Cancer in ctDNA during Sequential Androgen Receptor Pathway Inhibition. Clin Cancer Res 27:4610–4623, 2021 [DOI] [PubMed] [Google Scholar]
- 33.Kwan EM, Thangasamy IA, Teh J, et al. : Navigating systemic therapy for metastatic castration-naive prostate cancer. World J Urol 39:339–348, 2021 [DOI] [PubMed] [Google Scholar]
- 34.Koshkin VS, Patel VG, Ali A, et al. : PROMISE: a real-world clinical-genomic database to address knowledge gaps in prostate cancer. Prostate Cancer Prostatic Dis 25:388–396, 2022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Saad F, Clarke NW, Oya M, et al. : Olaparib plus abiraterone versus placebo plus abiraterone in metastatic castration-resistant prostate cancer (PROpel): final prespecified overall survival results of a randomised, double-blind, phase 3 trial. Lancet Oncol 24:1094–1108, 2023 [DOI] [PubMed] [Google Scholar]
- 36.Agarwal N, Azad AA, Carles J, et al. : Talazoparib plus enzalutamide in men with first-line metastatic castration-resistant prostate cancer (TALAPRO-2): a randomised, placebo-controlled, phase 3 trial. Lancet 402:291–303, 2023 [DOI] [PubMed] [Google Scholar]
- 37.Chi KN, Sandhu S, Smith MR, et al. : Niraparib plus abiraterone acetate with prednisone in patients with metastatic castration-resistant prostate cancer and homologous recombination repair gene alterations: second interim analysis of the randomized phase III MAGNITUDE trial. Ann Oncol 34:772–782, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Hwang C, Henderson NC, Chu SC, et al. : Biomarker-Directed Therapy in Black and White Men With Metastatic Castration-Resistant Prostate Cancer. JAMA Netw Open 6:e2334208, 2023 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 3. A Venn diagram of the overlap between HRD (BRCA1/2 and non-BRCA HRR alterations), MSI-hi/MMRd, and TMB.
Supplemental Figure 4. Swimmer’s plot showing specimen collection (black dot) for NGS testing in relation to disease state and therapies (PARP inhibitors, platinum chemotherapy, taxane chemotherapy, and immunotherapy). Pertinent findings on NGS testing are annotated on the side and color-coded by BRCA1/2 alterations (black), non-BRCA homologous recombination repair (HRR) gene alterations (gold), and mismatch repair deficient (MMRd) or microsatellite instability (MSI-hi; green). Bolded findings indicate those found on the second NGS test. Boxed are patients who received a targeted therapy (PARP inhibitor or immunotherapy) after a second NGS test.
Supplemental Figure 1. The number of days between each serial NGS test result.
Supplemental Figure 2. Sankey flow diagram showing disease state at the time of specimen collection for patients in the serial Next Generation Sequencing (NGS) group. Disease states include metastatic castration-resistant prostate cancer (mCRPC), non-metastatic castration-resistant prostate cancer (nmCRPC), metastatic hormone-sensitive prostate cancer (mHSPC), and non-metastatic hormone-sensitive prostate cancer (nmHSPC).