Abstract
Mismatch repair (MMR) gene mutations are rare in prostate cancer, and their histological and clinical characteristics are largely unknown. We conducted a retrospective study to explore disease characteristics and treatment outcomes of men with metastatic prostate cancer harboring germline and/or somatic MMR mutations detected using clinical-grade genomic assays. Thirteen patients with a deleterious MMR gene mutation were identified. Median age was 64 yr, 75% had grade group 5 (Gleason sum 9 or 10), 23% had intraductal histology, 46% had metastatic disease at initial diagnosis, and 31% had visceral metastases. Most patients (46%) had MSH6 mutations, 73% demonstrated microsatellite instability, and median tumor mutational load was 18/Mb (range, 3–165 mutations/Mb). Surprisingly, responses to standard hormonal therapies were very durable (median progression-free survival [PFS] of 67 mo to initial androgen deprivation and median PFS of 26 mo to abiraterone/enzalutamide). Two of four men receiving PD-1 inhibitors achieved a ≥50% prostate-specific antigen response at 12 wk, with a median PFS duration in these four men of 9 mo. Despite aggressive clinical and pathological features, patients with MMR-mutated advanced prostate cancer appear to have particular sensitivity to hormonal therapies, as well as anecdotal responses to PD-1 inhibitors. Certain histological features (grade group 5, intraductal carcinoma) should prompt evaluation for MMR deficiency. These data are only hypothesis generating.
Keywords: Prostate cancer, Mismatch repair, Microsatellite instability, MSH2, MSH6
Patient summary:
Prostate cancers with mismatch repair gene mutations have aggressive clinical and pathological features; however, these are very sensitive to standard and novel hormonal therapies, and also demonstrate anecdotal sensitivity to PD-1 inhibitors such as pembrolizumab.
Owing to the recent Food and Drug Administration’s approval of the PD-1 inhibitor pembrolizumab for the treatment of DNA mismatch repair–deficient (dMMR) or microsatellite instability-high (MSI-H) cancers of any histology [1], there has been a renewed interest in identifying tumors with these genomic features to aid therapy selection. However, while MMR deficiency and microsatellite instability are common features of gastrointestinal cancers, MMR gene mutations are rare in prostate cancer, estimated at 2– 5% of cases [2,3]. Owing to this low prevalence, data are lacking on the clinical and pathological characteristics of dMMR prostate cancer, and even less is known about the natural history and sensitivity to standard therapies for these cancers. Furthermore, except for isolated case reports (four patients, total) [3–5], there is no documented literature on the responsiveness of dMMR prostate cancers to immune-checkpoint inhibitors.
Here, we aimed to elucidate the clinical and histological features of prostate cancers harboring deleterious MMR gene mutations, with particular attention to potential characteristics that may alert a clinician to consider MMR/MSI testing. We also aimed at describing the sensitivity of MMR-mutated advanced prostate cancers to standard systemic therapies including androgen deprivation, novel hormonal therapies (abiraterone and enzalutamide), chemotherapy (docetaxel), as well as PD-1 inhibitor treatment. We demonstrate that these patients appear to be very sensitive to hormonal therapies but not to chemotherapy and that anecdotal benefits are also observed with PD-1 blockade.
We retrospectively queried our somatic genomic database at Johns Hopkins for prostate cancer cases with pathogenic loss-of-function (ie, inactivating) MMR mutations, and our germline genetic database for similar inherited MMR mutations. Both databases comprise recurrent and/or metastatic prostate cancer cases. Genes of interest included MSH2, MSH6, MLH1, and PMS2: the canonical MMR genes [6]. Predicted pathogenic mutations were defined using strict criteria: only protein-truncating mutations (frameshift, nonsense, or splicing lesions) as well as genomic deletions or structural rearrangements were considered deleterious. Somatic next-generation tumor-DNA sequencing had previously been performed for clinical indications using the Personal Genome Diagnostics (PGDx, Baltimore, MD, USA) 125-gene targeted panel [7]. Germline genetic testing was also performed previously for clinical reasons, using the saliva-based 30-gene targeted next-generation panel offered by Color Genomics (Burlingame, CA, USA) [8]. Where available, primary or metastatic tumor tissue was used to perform standard immunohistochemical (IHC) analysis for the detection of the four MMR proteins. Clinical outcomes of a variety of systemic therapies were coded according to the PCWG3 criteria [9]. Prostate-specific antigen (PSA) response evaluation required ≥12 wk of follow-up, and the 12-wk PSA response rate is reported.
Thirteen metastatic prostate cancer patients with pathogenic MMR gene mutations were identified: 10 from screening 236 somatic sequencing results (4.2%) and three from screening 348 germline sequencing results (0.9%). Table 1 shows their baseline characteristics. Median age was 64 yr, 69% were white, 75% had grade group 5 (Gleason sum 9 or 10), 23% had intraductal histology, 46% had metastases at initial diagnosis, and 31% had visceral involvement. All patients had received standard androgen deprivation therapy (ADT), 46% (6/13) had received first-line abiraterone or enzalutamide, 15% (2/13) had received docetaxel, and 31% (4/13) had received PD-1 blockade.
Table 1 –
Baseline demographic and disease characteristics of our 13 MMR-deficient prostate cancers
| Characteristic | MMR-deficient men (N = 13) | MMR-proficient men (N = 114) |
|---|---|---|
| Age at diagnosis (yr) | ||
| Median (Q1–Q3) | 64 (61–70) | 63 (59–69) |
| Race, N (%) | ||
| White | 9 (69) | 99 (87) |
| Presence of any secondary malignancy, N (%) | 3 (23) | 9 (8) |
| Family history of cancer, N (%) | ||
| First-degree relative | 8 (62) | 59 (52) |
| Non–first-degree relative | 5 (38) | 18 (16) |
| Gleason sum at diagnosis, N (%) | ||
| ≤7 | 2 (15) | 30 (26) |
| ≥8 | 10 (77) | 77 (67) |
| Unknown | 1 (8) | 7 (6) |
| Presence of perineural invasion, N (%) | 4 (31) | 68 (60) |
| Presence of variant histology, N (%) | ||
| Ductal/intraductal | 3 (23) | 14 (12) |
| Neuroendocrine | 1 (8) | 0 (0) |
| Tumor stage at diagnosis, N (%) | ||
| T1/T2 | 3 (23) | 36 (32) |
| T3/T4 | 10 (77) | 78 (68) |
| Lymph node stage at diagnosis, N (%) | ||
| N1 | 5 (38) | 15 (13) |
| Metastatic stage at diagnosis, N (%) | ||
| M1 | 6 (46) | 37 (32) |
| Presence of bone metastasis, N (%) | ||
| Bone only | 3 (23) | 34 (30) |
| With visceral metastasis (lung, liver) | 4 (31) | 22 (19) |
| Presence of lung metastasis only, N (%) | 2 (15) | 4 (3) |
| Presence of liver metastasis, N (%) | 2 (15) | 10 (9) |
| Use of standard ADT, N (%) | 13 (100) | 114 (100) |
| Use of abiraterone, N (%) | 3 (23) | 47 (41) |
| Use of enzalutamide, N (%) | 5 (38) | 28 (25) |
| Use of docetaxel, N (%) | 2 (15) | 38 (33) |
| Use of PD-1 inhibitor, N (%) | 4 (31) | 2 (2) |
| PSA at diagnosis (ng/ml) | ||
| Median (Q1–Q3) | 10 (5.4–43) | 13 (5.5–32) |
ADT = androgen deprivation therapy; MMR = mismatch repair; PSA = prostate-specific antigen.
For comparison, we also include baseline characteristics for a group of 114 MMR-proficient men from our somatic sequencing database with full clinical and outcome data.
Table 2 summarizes the genomic characteristics. Two patients with germline MMR mutations did not have adequate tumor tissue available for somatic DNA analyses or IHC studies, and one additional patient did not have available tissue for IHC studies only. Most men had MSH6 (46%; 6/13) or MSH2 mutations (23%; 3/13), and median tumor mutational burden was 18 mutations/Mb (range, 3–165 mutations/Mb). Of those with adequate tissue available for sequencing, 73% (8/11) demonstrated microsatellite instability: 27% (3/11) had no MSI markers shifted, 36% (4/11) had one to two markers shifted, and 36% (4/11) had three to four markers shifted. Median mutational loads were 21 and 6 mutations/Mb for MSI-positive and MSI-negative patients, respectively. While two of three patients with microsatellite-stable status (#2 and #7; both with PMS2 mutations) had intact MMR protein expression, the third patient (#1) demonstrated loss of MSH2 and MSH6 proteins by IHC consistent with genomic MSH2 inactivation. Notably, 64% of patients (7/11) had coexisting TP53 mutations and 36% (4/11) had TMPRSS2-ERG fusions.
Table 2 –
Clinicopathological and molecular characteristics of patients with pathogenic MMR gene mutations
| Patient ID | Gleason score, tumor stage | Specimen type tested | Variant histology | MMR gene mutation | Protein IHC status | MSI markers shifted a | MSI status a | Mutation load | Other mutations of interest |
|---|---|---|---|---|---|---|---|---|---|
| #1 | 4 + 5 = 9 | RP | None noted | MSH2 (C778X*) | MSH2 and MSH6 loss b | 0/5 | MSS | 11 muts/Mb c | AKT1 (E17K) |
| T3a N0 M0 | MLH1 and PMS2 intact | CTNNB1 (D32G) | |||||||
| TMPRSS2-ERG fusion | |||||||||
| #2 | 3 + 4 = 7 | RP | None noted | PMS2 (L729Qfs*6) | MSH2, MSH6, MLH1, PMS2 all intact | 0/5 | MSS | 3 muts/Mb | TP53 (R273H) |
| T3bN1 M0 | PMS2 (T728A) | ||||||||
| #3 | 3 + 4 = 7 | RP | None noted | gMSH6 (A1320Sfs*5) | Adequate tissue not available | No somatic (tumor) DNA analysis was performed | |||
| T3b N0 M0 | |||||||||
| #4 | 5 + 5 = 10 | Bx | None noted | MSH6 (F1088Sfs*2) | MSH6 loss only | 3/5 | MSI-high | 18 muts/Mb | PMS2 (D414Tfs*34) |
| MSH2, MLH1, PMS2 intact | JAK1 (N339Ifs*3) | ||||||||
| RET (L1048Sfs*61) | |||||||||
| RNF43 (G659Vfs*41) | |||||||||
| #5 | 4 + 5 = 9 | Bx | None noted | MSH6 (F1088Lfs*5) | MSH2 and MSH6 loss | 3/5 | MSI-high | 35 muts/Mb | BRCA2 (N1784Kfs*3) |
| MLH1 and PMS2 intact | HRAS (P167Rfs*51) | ||||||||
| JAK2 (N457Mfs*22) | |||||||||
| TP53 (D281N) | |||||||||
| #6 | 4 + 5 = 9 | Bx | Intraductal carcinoma | gMSH6 (V1192Lfs*3) | Adequate tissue not available | No somatic (tumor) DNA analysis was performed | |||
| #7 | 4 + 5 = 9 | RP | None noted | PMS2 (M622Efs*5) | MSH2, MSH6, MLH1, PMS2 all intact | 0/5 | MSS | 6 muts/Mb | KMT2A (S774Vfs*12) |
| T3b N0 M0 | TP53 (H179Q) | ||||||||
| #8 | 4 + 5 = 9 | RP | None noted | MLH1 (heterozygous gene deletion) | MLH1 and PMS2 loss | 2/5 | MSI-high | 13 muts/Mb | PTEN (K267Efs*9) |
| T3a NO M0 | MSH2 and MSH6 intact | RNF43 (G659Vfs*41) | |||||||
| TP53 (T155I) | |||||||||
| TMPRSS2-ERG fusion | |||||||||
| #9 | Unknown (no primary tumor biopsy) | Lymph node | None noted | MSH2 (L376Ffs*13) | MSH2 and MSH6 loss | 4/5 | MSI-high | 42 muts/Mb | PMS1 (T256Hfs*2) |
| MLH1 and PMS2 intact | TP53 (Q167X*) | ||||||||
| TP53 (S240G) | |||||||||
| PIK3CA (H1047R) | |||||||||
| #10 | 4 + 5 = 9 | Bx | Intraductal carcinoma | MSH6 (E192X*) | Adequate tissue not available | 1/5 | MSI-low | 8 muts/Mb | TP53 (E271V) |
| BRCA2 (P3189H) | |||||||||
| #11 | 4 + 5 = 9 | Bx | None noted | MLH1 (T206Mfs*23) | PMS2 loss only | 2/5 | MSI-high | 20 muts/Mb | BRCA1 (Q1111Efs*5) |
| MLH-1, MSH2, MSH6 intact | PTEN (T319Ifs*1) | ||||||||
| RNF43 (G659Vfs*41) | |||||||||
| CTNNB1 (T41A) | |||||||||
| TMPRSS2-ERG fusion | |||||||||
| #12 | 4 + 4 = 8 | Bx | None noted | gMSH6 (E230Sfs*4) | MSH6 loss only | 2/5 | MSI-high | 22 muts/Mb | TP53 (A76Vfs*55) |
| MSH2, MLH, and PMS2 all intact | TMPRSS2-ERG fusion | ||||||||
| #13 | 4 + 5 = 9 | RP | Intraductal carcinoma | MSH2 (E809X*) + LOH of 2nd allele | MSH2 and MSH6 loss | 4/5 | MSI-high | 165 muts/Mb | MSH6 (F1104Lfs*11) |
| T3a N0 M0 | MLH1 and PMS2 intact | ATM (L663Ffs*2) | |||||||
| ERCC4 (M361Nfs*4) | |||||||||
| ERCC5 (E474Nfs*15) | |||||||||
| FANCM (V1336Lfs*2) | |||||||||
Bx = prostate biopsy; g = germline mutation; IHC = immunohistochemistry; MMR = mismatch repair; MSI = microsatellite instability; MSS = microsatellite stable; muts = mutations; RP = radical prostatectomy.
MSI status was determined from targeted next-generation DNA sequencing (Personal Genome Diagnostics Baltimore, MD, USA) using the five well-characterized NIH-defined mononucleotide sequences (BAT-25, BAT-26, NR-21, NR-24, and MONO-27), as previously described [7]. MSI-high status is defined by shifts in two to five markers, MSI-low status is defined by a shift in one marker, and MSS status is defined by no shifted markers.
In cases of protein loss by IHC, the loss was typically homogeneous rather than focal for all of the MMR proteins assayed.
Tumors with mutational loads of ≥10 mutations/Mb were considered hypermutated.
Despite aggressive clinicopathological features as well as frequent TP53 mutations, MMR-deficient patients demonstrated high sensitivity to hormonal therapies. All 13 men received standard ADT (without concurrent docetaxel or abiraterone) as initial systemic therapy for metastatic disease, and 85% achieved a >90% PSA response rate (11/13; median PSA reduction, 99%), with median PSA progression-free survival (PFS) of 55 (95% confidence interval [CI] 50–73) mo and median PFS of 66 (95% CI 55–77) mo. (By comparison, median PFS to first-line ADT among 114 MMR-proficient men from our somatic sequencing database with full clinical data was 27 [95% CI 22–32] mo.) Sensitivity to first-line abiraterone or enzalutamide was also high among the six MMR-deficient patients evaluable for this outcome, of whom 83% (five of six patients) achieved a >50% PSA response (median PSA reduction, 80%), with median PSA PFS of 24 (95%CI 5–not reached) mo and median PFS of 26 (95% CI 6–not reached) mo. (By comparison, median PFS to first-line abiraterone/enzalutamide among 75 MMR-proficient men from our somatic sequencing database with full clinical information was 12 [95% CI 10–14] mo.) Two MMR-deficient patients received docetaxel, of whom one achieved a >50% PSA response (median PSA reduction, 41%), with median PSA PFS of 6 (95% CI 4–9) mo and median PFS of 7 (95% CI 5–10) mo. Finally, four patients (#6, #8, #9, and #10) received PD-1 inhibitor treatment as fourth- to sixth-line systemic therapy: two using nivolumab and two using pembrolizumab. Half of patients (two of four patients [#9 and #10]; both PD-L1 positive by IHC) achieved a >50% PSA response (median PSA reduction, 56%), with median PSA PFS of 7 (95% CI 3–9) mo and median PFS of 9 (95% CI 4–11) mo. Three of these patients (75% [#8, #9, and #10]) also achieved an objective soft-tissue response lasting for 3–9+ months.
Mismatch repair-deficient prostate cancers are rare, representing <5% of tumors [2,3], but their detection has therapeutic implications [1,3,6]. We report the first case series of MMR-mutated prostate cancers by intersecting data from our institutional genomic and clinical databases. As exemplified here and in our prior studies, diagnosis of dMMR prostate cancers can be challenging because: (1) not all MMR mutations (even those predicted to be inactivating [frameshift, nonsense, and splicing lesions]) result in MMR protein loss or microsatellite instability [10], (2) the five NIH-defined microsatellite loci may be inadequate in detecting true microsatellite instability in prostate cancer and a more expanded microsatellite panel might increase sensitivity [11], (3) not all inactivating MMR gene mutations (especially structural rearrangements) can be detected by clinical-grade targeted-exon sequencing [5,12], and (4) not all dMMR prostate cancers demonstrate hypermutation or dense CD8 T-cell infiltrates [5,10]. Importantly, particular tumor grades (grade group 5, primary pattern 5) [10] or variant histologies (intraductal/ductal carcinoma and small cell carcinoma) [5,8,10] may enrich for MMR alterations.
Clinically, we have shown that dMMR prostate cancers may demonstrate remarkable sensitivity to standard ADT as well as novel hormonal agents, despite a high prevalence of TP53 mutations [13]. PSA responses as well as median duration of responses to both conventional ADT and first-line abiraterone/enzalutamide far exceeded historical estimates [14], and were greater than those seen in our MMR-proficient population. These data raise the question of whether hormonal therapies may be immunomodulatory, potentially contributing to their efficacy in the context of dMMR prostate cancer [15]. Conversely, responses to docetaxel appeared modest, although interpretations are limited by the very small number of taxane-treated patients. Finally, while anecdotal PSA and objective responses were observed using PD-1 inhibitors, durability of such responses (median, 9 mo) could be improved by using these agents earlier or in combination with other immunological or standard therapies [1]. Our conclusions are limited by the small cohort size and the retrospective nature of this analysis, and should be interpreted with extreme caution.
Acknowledgments
Funding/Support and role of the sponsor: This work was partially supported by National Institutes of Health Cancer Center Support Grant P30 CA006973 (E.S.A. and T.L.L.), Department of Defense grant W81XWH-16-PCRP-CCRSA (E.S.A.), and the Bloomberg-Kimmel Institute for Cancer Immunotherapy (E.S.A., E.S., and D.M.P.).
Footnotes
Financial disclosures: Emmanuel S. Antonarakis certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Emmanuel S. Antonarakis is a paid consultant/advisor to Janssen, Astellas, Sanofi, Dendreon, Medivation, ESSA, AstraZeneca, Clovis, and Merck; he has received research funding to his institution from Janssen, Johnson & Johnson, Sanofi, Dendreon, Genentech, Novartis, Tokai, Bristol Myers-Squibb, AstraZeneca, Clovis, and Merck; and he is the co-inventor of a biomarker technology that has been licensed to Qiagen. Tamara L. Lotan is a paid consultant/advisor to Janssen and has received research funding from Ventana/Roche.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- [1].Isaacsson Velho P, Antonarakis ES. PD-1/PD-L1 pathway inhibitors in advanced prostate cancer. Expert Rev Clin Pharmacol 2018;11:475–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Robinson D, Van Allen EM, Wu YM, et al. Integrative clinical genomics of advanced prostate cancer. Cell 2015;161:1215–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [3].Le DT, Durham JN, Smith KN, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 2017;357:409–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Graff JN, Alumkal JJ, Drake CG, et al. Early evidence of anti-PD-1 activity in enzalutamide-resistant prostate cancer. Oncotarget 2016;7:52810–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Schweizer MT, Cheng HH, Tretiakova MS, et al. Mismatch repair deficiency may be common in ductal adenocarcinoma of the prostate. Oncotarget 2016;7:82504–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Lee V, Murphy A, Le DT, Diaz LA Jr. Mismatch repair deficiency and response to immune checkpoint blockade. Oncologist 2016;21:1200–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Boudadi K, Suzman DL, Anagnostou V, et al. Ipilimumab plus nivolumab and DNA-repair defects in AR-V7–expressing metastatic prostate cancer. Oncotarget 2018;9:28561–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Isaacsson Velho P, Silberstein JL, Markowski MC, et al. Intraductal/ductal histology and lymphovascular invasion are associated with germline DNA-repair gene mutations in prostate cancer. Prostate 2018;78:401–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Scher HI, Morris MJ, Stadler WM, et al. Trial design and objectives for castrationresistant prostate cancer: Updated recommendations from the Prostate Cancer Clinical Trials Working Group 3 (PCWG3). J Clin Oncol 2016;34:1402–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Guedes LB, Antonarakis ES, Schweizer MT, et al. MSH2 loss in primary prostate cancer. Clin Cancer Res 2017;23:6863–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Hempelmann JA, Lockwood CM, Konnick EQ, et al. Microsatellite instability in prostate cancer by PCR or next-generation sequencing. J Immunother Cancer 2018;6:29–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [12].Pritchard CC, Morrissey C, Kumar A, et al. Complex MSH2 and MSH6 mutations in hypermutated microsatellite unstable advanced prostate cancer. Nat Commun 2014;5:4988–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Maughan BL, Guedes LB, Boucher K, et al. p53 status in the primary tumor predicts efficacy of subsequent abiraterone and enzalutamide in castration-resistant prostate cancer. Prostate Cancer Prostatic Dis 2018;21:260–8. [DOI] [PubMed] [Google Scholar]
- [14].Antonarakis ES, Eisenberger MA. Expanding treatment options for metastatic prostate cancer. N Engl J Med 2011;364:2055–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].Mercader M, Bodner BK, Moser MT, et al. T cell infiltration of the prostate induced by androgen withdrawal in patients with prostate cancer. Proc Natl Acad Sci U S A 2001;98:14565–70. [DOI] [PMC free article] [PubMed] [Google Scholar]