- Email: [email protected]
Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer
EUO-262; No. of Pages 5 E U RO P E A N U RO L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
available at www.sciencedirect.com journal homepage: euoncology.europeanurology.com
Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer Heather J. Chalfin *, Stephanie A. Glavaris, Michael A. Gorin, Max R. Kates, Megan H. Fong, Liang Dong, Andres Matoso, Trinity J. Bivalacqua, Michael H. Johnson, Kenneth J. Pienta, Noah M. Hahn, David J. McConkey The James Buchanan Brady Urological Institute and Greenberg Bladder Cancer Institute, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Article info
Abstract
Article history: Received 8 April 2019 Received in revised form 2 July 2019 Accepted August 15, 2019
Despite considerable advances in the management of urothelial carcinoma (UC), better risk stratification and enhanced detection of minimal residual disease are still urgent priorities to prolong survival while avoiding the morbidity of overtreatment. Circulating tumor cells and DNA (CTCs, ctDNA) are two biologically distinct “liquid biopsies” that may potentially address this need, although they have been understudied in UC to date and their relative utility is unknown. To this end, matched CTC and ctDNA samples were collected for a head-to-head comparison in a pilot study of 16 patients with metastatic UC. CTCs were defined as cytokeratin- and/or EpCAM-positive using the RareCyte direct imaging platform. ctDNA was assayed using the PlasmaSelect64 probe-capture assay. 75% of patients had detectable CTCs, and 73% had detectable somatic mutations, with no correlation between CTC count and ctDNA. 91% of patients had tissue confirmation of at least one plasma mutation and, importantly, several clinically actionable mutations were detected in plasma that were not found in the matching tumor. A ctDNA fraction of >2% was significantly associated with worse overall survival (p = 0.039) whereas CTC detection was not (p = 0.46). Notably, using a predefined gene panel for ctDNA detection had a high but not complete detection rate in metastatic UC, similar to what has been described for a custom tissue-personalized assay approach. In sum, both liquid biopsies show promise in UC and deserve further investigation. Patient summary: New “liquid biopsy” blood tests are emerging for urothelial cancer aimed at early detection and avoiding overtreatment. Our results suggest that two such tests provide complementary information: circulating tumor cells may be best for studying the biological features of a person’s cancer, whereas circulating tumor DNA may be better for early detection.
Associate Editor: Ashish Kamat Keywords: Bladder cancer Liquid biopsy Circulating tumor cells Circulating tumor DNA
Published by Elsevier B.V. on behalf of European Association of Urology.
* Corresponding author. The James Buchanan Brady Urological Institute, Johns Hopkins University School of Medicine, 600 N. Wolfe Street, Baltimore, MD 21287, USA. Tel.: +1 443 7423130, Fax: +1 410 9550833. E-mail address: [email protected] (H.J. Chalfin).
https://doi.org/10.1016/j.euo.2019.08.004 2588-9311/Published by Elsevier B.V. on behalf of European Association of Urology.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 2
E U R O P E A N U R O L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
Urothelial carcinoma (UC) of the bladder is a common urologic malignancy, with an estimated 79 000 new cases diagnosed each year in the USA [1]. Despite considerable advances in the treatment of UC, it remains an aggressive disease and the gold standard of cystectomy is a morbid procedure. Approximately 40% of patients will enjoy a complete response to neoadjuvant chemotherapy with no residual tumor found at the time of cystectomy; however, these cases cannot be reliably identified before surgery with any currently available test [2]. Better detection of minimal residual disease and better risk stratification are an urgent priority to minimize overtreatment and prolong survival. Many minimally invasive “liquid biopsy” technologies are currently under investigation to address this need, including circulating tumor cells (CTCs). These are malignant cancer cells that can be detected in peripheral blood. CTC morphology, protein expression, and genomics can be assessed at the
single cell level, and thus CTC assays are not simply an interchangeable alternative for the more widely studied liquid biopsy of circulating tumor DNA (ctDNA). Furthermore, CTC detection reveals that an intact cell has survived in the peripheral circulation, which is not biologically equivalent to detection of fragments of shed tumor DNA via a ctDNA assay. To date, CTCs and ctDNA have been understudied in UC. The vast majority of the work investigating CTCs in bladder cancer has used the CellSearch test, with no consensus on their prognostic significance [3]. These discordant observations are likely because of the limitations of the CellSearch technology, which cannot detect epithelial marker– negative CTCs [4]. Furthermore, the emphasis has been on counting CTCs as opposed to harnessing the opportunity to study cell morphology and biomarkers at the single cell level. More recently, novel selection-free platforms with enhanced detection capabilities and a greater focus on
Fig. 1 – Individual patient CTC and ctDNA results grouped by ctDNA fraction detected. Patients are grouped by high (red box around liquid biopsy time point) and low (green box) ctDNA fraction. One patient did not have ctDNA results (black box). The ctDNA fraction is shown in red/green, and the CTC count displayed below in black. Blue boxes represent tumors that were sequenced. Patient 2 had basal cell skin cancer resected. Patient 3 had seminoma treated with orchiectomy and radiation at age 22 yr. Patient 4 had squamous cell skin cancer resected. Patients 8 and 13 had Gleason 6 prostate cancer (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). CHEMO = chemotherapy; EBRT = external beam radiation therapy; LNU = left nephroureterectomy; MET = metastatic; NAC = neoadjuvant chemotherapy; NephroU = nephroureterectomy; NMIBC = non–muscle-invasive bladder cancer; RC = radical cystectomy; TURBT = transurethral resection of bladder tumor.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 E U R O P E A N U RO L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
morphology have started to emerge [5]. For ctDNA, there is also no consensus on the optimal strategy for detection. A personalized medicine approach of unique droplet digital polymerase chain reaction (ddPCR) assays targeted to each individual patient’s known mutations should offer greater sensitivity than a generic next-generation sequencing (NGS) panel. Conversely, an NGS panel may be able to detect mutations that have arisen later and are not represented in the primary tumor. Compounding these considerations is the distinctly lower expression of CTCs and ctDNA in UC than in other malignancies [6]. Given the potential of these distinct liquid biopsies for risk stratification and the lack of information on their relative utility in UC, we performed a head-to-head comparison of a selection-free immunofluorescence-based CTC assay (RareCyte) and a plasma NGS panel for ctDNA detection in a pilot cohort of 16 consecutive patients with metastatic UC. Fig. 1 shows their clinical course and treatment. We found similar detection rates for CTCs (75% of patients; median 2.5 CTCs, range 0–170) and ctDNA (73% of patients; median of 2 somatic mutations, range 0–7) yet interestingly the CTC count did not
3
correlate with cell-free DNA or the ctDNA fraction (Supplementary Fig. 1). We discovered a similar genomic landscape to that previously described, with frequent mutations in TP53, TERT, and ERBB2 (Fig. 2) [7]. With a median follow up of 5.9 mo, a ctDNA fraction of >2% was associated with worse overall survival (OS; p = 0.039) with a median OS of 38 wk versus not reached for ctDNA 2% (Supplementary Fig. 2). By contrast, CTC detection did not have a significant impact on OS (p = 0.46), with a median OS of 59 wk for no CTCs versus 25 wk for 1 CTC (Supplementary Fig. 3). The cutoffs were defined a priori on the basis of previous work [7]. Notably, our ctDNA detection rate of 73% is in line with that of three other groups that reported 65%, 69%, and 73% for small cohorts [8]. Overall, these numbers are disappointing, as they represent incomplete detection in advanced disease, and the ideal liquid biopsy should be sensitive enough for early diagnosis and detection of minimal residual disease. Of note, there is one commercially available NGS ctDNA panel (Guardant360) with a detection rate of 90% reported for genomic alterations in advanced UC; however, the same group showed that the results for
Fig. 2 – Mutational landscape of metastatic bladder cancer plasma samples. Each column represents one plasma sample for a given patient, with the corresponding circulating tumor DNA fraction plotted on the bar graph above. Each mutated gene is shown on a row, with the mutation frequency of that gene in the overall cohort shown on the bar graph to the right. Boxes are color coded with respect to the mutant allele fraction of the mutation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 4
E U R O P E A N U R O L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
this panel were significantly discordant with tumor collected at the same time [9–11]. In the present study we found higher mutant allele frequencies (MAFs) for plasma mutations that were concordant with tissue (11.16 vs 0.40; p 0001) and that were triplicated in replicates (43.38 vs 21.09; p = 0.07), suggesting a certain threshold for detection with the NGS assay. Logically, it might be expected that a tissue-personalized ddPCR assay would provide greater sensitivity; however, this approach also has limitations due to tumor heterogeneity and clonal evolution. In this context, in the present study several potentially clinically actionable mutations were only detected in plasma and not in matched tumor specimens collected at various time points relative to the liquid biopsies (91% of patients had tissue confirmation of at least one plasma mutation, Supplementary Figs. 4 and 5; all available primary/metastatic tumor was sequenced, Supplementary Fig. 1). The main example of a tissue-personalized ddPCR assay in UC was described by BirkenkampDemtröder et al [12]. Five of 12 patients (42%) had at least one undetectable result during metastatic disease, with one patient having seven negative reads. These patients received heterogeneous treatments, and the question of sensitivity for the metastatic setting of ddPCR assays designed based on primary tumor alterations deserves further dedicated study. An additional challenge for NGS assays is that quantification of mutant genome expression must be extrapolated from the MAFs. By design, MAF is affected by the overall coverage of the region of interest, which is not uniform across the genome. This limitation affected the present study, and it must be emphasized that our analysis was certainly impacted by small numbers and heterogeneous clinicopathologic factors at baseline. CTC detection was not significantly associated with OS; however, this could simply reflect a lack of statistical power necessary to detect a relationship that does exist. Despite these challenges, liquid biopsy represents the next frontier for diagnosis, management, and follow-up for solid malignancies. Our results suggest that CTCs and ctDNA can provide complementary information: ctDNA quantification was relevant for survival and the NGS assay detected individual patient mutations that were not identified in tumor, whereas the CTC platform demonstrated feasibility for capture of immunoflourescent images revealing the morphology of circulating malignant cells as well as protein expression at the single cell level. When considering the ideal CTC assay, the emphasis should shift from simple enumeration to a biological assay. It has been demonstrated in preclinical studies that the CTC count fluctuates from 0 to 54 in a 5-min time span [13], and in the present study we measured median fluctuations from moment to moment and day to day of 2 and 7 CTCs, respectively. Using the Epic CTC platform, Scher et al [5] showed that the degree of cell phenotype heterogeneity in prostate cancer can inform treatment response to taxane versus androgen inhibitor therapy, and similar studies in UC are certainly warranted. Finally, we have shown that quantitative measures of ctDNA may be used for patient risk stratification, although qualitative mutation results can also have important implications for therapy decisions. We have
described the unique challenge of ctDNA detection in UC and noted that the optimal approach must provide sensitivity without losing the ability to detect clonal evolution. Author contributions: Heather J. Chalfin had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Chalfin, Glavaris, Gorin, Hahn, McConkey. Acquisition of data: Chalfin, Glavaris, Kates, Fong, Dong, Matoso, Bivalacqua, Johnson, Pienta, Hahn, McConkey. Analysis and interpretation of data: Chalfin, Glavaris, Gorin, Dong, Hahn, McConkey. Drafting of the manuscript: Chalfin, Glavaris. Critical revision of the manuscript for important intellectual content: Chalfin, Glavaris, Gorin, Kates, Bivalacqua, Pienta, Hahn, McConkey. Statistical analysis: Chalfin, Glavaris, Gorin. Obtaining funding: Chalfin, Hahn, McConkey. Administrative, technical, or material support: Glavaris, Fong, Dong, Matoso. Supervision: Gorin, Kates, Bivalacqua, Johnson, Pienta, Hahn, McConkey. Other: None. Financial disclosures: Heather J. Chalfin 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: Kenneth J. Pienta has acted as a consultant for Celsee Diagnostics and acts as the Chief Medical Officer for Cuebiophrma. The PlasmaSELECT test is owned by Personal Genome Diagnostics (PGDx). The Johns Hopkins University has licensed technologies to PGDx and, as a result, has a financial interest in the company. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies. The remaining authors have nothing to disclose. Funding/Support and role of the sponsor: H.J.C. is supported by a grant from the Johns Hopkins Greenberg Bladder Cancer Institute. The sponsor played a role in review of the manuscript.
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j. euo.2019.08.004.
References [1] Howlader N, Noone AM, Krapcho M, editors. SEER cancer statistics review, 1975–2013. Bethesda, MD: National Cancer Institute; 2016. http://seer.cancer.gov/csr/1975_2013/ [2] deVere White RW, Lara Jr PN, Goldman B, et al. A sequential treatment approach tomyoinvasive urothelial cancer: a phase II Southwest Oncology Group trial (S0219). J Urol 2009;181:2476–80. [3] Gallagher DJ, Milowsky MI, Ishill N, et al. Detection of circulating tumor cells in patients with urothelial cancer. Ann Oncol 2009;20:305–8. [4] McDaniel AS, Ferraldeschi R, Krupa R, et al. Phenotypic diversity of circulating tumour cells in patients with metastatic castrationresistant prostate cancer. BJU Int 2017;120:E30–44.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 E U R O P E A N U RO L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
5
[5] Scher HI, Graf RP, Schreiber NA, et al. Phenotypic heterogeneity of
[10] Barata PC, Koshin VS, Funchain P, et al. Next-generation sequencing
circulating tumor cells informs clinical decisions between AR sig-
(NGS) of cell-free circulating tumor DNA and tumor tissue in
naling inhibitors and taxanes in metastatic prostate cancer. Cancer
patients with advanced urothelial cancer: a pilot assessment of
Res 2017;77:5687–98. [6] Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating
concordance. Ann Oncol 2017;28:2458–63. [11] Grivas P., Lalani A.A., Pond G.R., et al. Circulating tumor DNA
tumor DNA in early- and late-stage human malignancies. Sci Transl
alterations in advanced urothelial carcinoma and association with
Med 2014;6:224ra24.
clinical outcomes: a pilot study. Eur Urol Oncol. In press. http://dx.
[7] Vandekerkhove G, Todenhöfer T, Annala M, et al. Circulating tumor DNA reveals clinically actionable somatic genome of metastatic bladder cancer. Clin Cancer Res 2017;23:6487–97. [8] Tan MP, Attard G, Huddart RA. Circulating tumour DNA in muscleinvasive bladder cancer. Int J Mol Sci 2018;19:2568. [9] Agarwal N, Pal SK, Hahn AW, et al. Characterization of metastatic
doi.org/10.1016/j.euo.2019.02.004. [12] Birkenkamp-Demtröder K, Christensen E, Nordentoft I, et al. Monitoring treatment response and metastatic relapse in advanced bladder cancer by liquid biopsy analysis. Eur Urol 2018;73:535–40. [13] Fei F, Du Y, Di G, Wu J, Shao Z. Are changes in circulating tumor cell (CTC) count associated with the response to neoadjuvant chemo-
urothelial carcinoma via comprehensive genomic profiling of cir-
therapy in local advanced breast cancer? A meta-analysis. Oncol Res
culating tumor DNA. Cancer 2018;124:2115–24.
Treat 2014;37:250–4.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
available at www.sciencedirect.com journal homepage: euoncology.europeanurology.com
Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer Heather J. Chalfin *, Stephanie A. Glavaris, Michael A. Gorin, Max R. Kates, Megan H. Fong, Liang Dong, Andres Matoso, Trinity J. Bivalacqua, Michael H. Johnson, Kenneth J. Pienta, Noah M. Hahn, David J. McConkey The James Buchanan Brady Urological Institute and Greenberg Bladder Cancer Institute, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Article info
Abstract
Article history: Received 8 April 2019 Received in revised form 2 July 2019 Accepted August 15, 2019
Despite considerable advances in the management of urothelial carcinoma (UC), better risk stratification and enhanced detection of minimal residual disease are still urgent priorities to prolong survival while avoiding the morbidity of overtreatment. Circulating tumor cells and DNA (CTCs, ctDNA) are two biologically distinct “liquid biopsies” that may potentially address this need, although they have been understudied in UC to date and their relative utility is unknown. To this end, matched CTC and ctDNA samples were collected for a head-to-head comparison in a pilot study of 16 patients with metastatic UC. CTCs were defined as cytokeratin- and/or EpCAM-positive using the RareCyte direct imaging platform. ctDNA was assayed using the PlasmaSelect64 probe-capture assay. 75% of patients had detectable CTCs, and 73% had detectable somatic mutations, with no correlation between CTC count and ctDNA. 91% of patients had tissue confirmation of at least one plasma mutation and, importantly, several clinically actionable mutations were detected in plasma that were not found in the matching tumor. A ctDNA fraction of >2% was significantly associated with worse overall survival (p = 0.039) whereas CTC detection was not (p = 0.46). Notably, using a predefined gene panel for ctDNA detection had a high but not complete detection rate in metastatic UC, similar to what has been described for a custom tissue-personalized assay approach. In sum, both liquid biopsies show promise in UC and deserve further investigation. Patient summary: New “liquid biopsy” blood tests are emerging for urothelial cancer aimed at early detection and avoiding overtreatment. Our results suggest that two such tests provide complementary information: circulating tumor cells may be best for studying the biological features of a person’s cancer, whereas circulating tumor DNA may be better for early detection.
Associate Editor: Ashish Kamat Keywords: Bladder cancer Liquid biopsy Circulating tumor cells Circulating tumor DNA
Published by Elsevier B.V. on behalf of European Association of Urology.
* Corresponding author. The James Buchanan Brady Urological Institute, Johns Hopkins University School of Medicine, 600 N. Wolfe Street, Baltimore, MD 21287, USA. Tel.: +1 443 7423130, Fax: +1 410 9550833. E-mail address: [email protected] (H.J. Chalfin).
https://doi.org/10.1016/j.euo.2019.08.004 2588-9311/Published by Elsevier B.V. on behalf of European Association of Urology.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 2
E U R O P E A N U R O L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
Urothelial carcinoma (UC) of the bladder is a common urologic malignancy, with an estimated 79 000 new cases diagnosed each year in the USA [1]. Despite considerable advances in the treatment of UC, it remains an aggressive disease and the gold standard of cystectomy is a morbid procedure. Approximately 40% of patients will enjoy a complete response to neoadjuvant chemotherapy with no residual tumor found at the time of cystectomy; however, these cases cannot be reliably identified before surgery with any currently available test [2]. Better detection of minimal residual disease and better risk stratification are an urgent priority to minimize overtreatment and prolong survival. Many minimally invasive “liquid biopsy” technologies are currently under investigation to address this need, including circulating tumor cells (CTCs). These are malignant cancer cells that can be detected in peripheral blood. CTC morphology, protein expression, and genomics can be assessed at the
single cell level, and thus CTC assays are not simply an interchangeable alternative for the more widely studied liquid biopsy of circulating tumor DNA (ctDNA). Furthermore, CTC detection reveals that an intact cell has survived in the peripheral circulation, which is not biologically equivalent to detection of fragments of shed tumor DNA via a ctDNA assay. To date, CTCs and ctDNA have been understudied in UC. The vast majority of the work investigating CTCs in bladder cancer has used the CellSearch test, with no consensus on their prognostic significance [3]. These discordant observations are likely because of the limitations of the CellSearch technology, which cannot detect epithelial marker– negative CTCs [4]. Furthermore, the emphasis has been on counting CTCs as opposed to harnessing the opportunity to study cell morphology and biomarkers at the single cell level. More recently, novel selection-free platforms with enhanced detection capabilities and a greater focus on
Fig. 1 – Individual patient CTC and ctDNA results grouped by ctDNA fraction detected. Patients are grouped by high (red box around liquid biopsy time point) and low (green box) ctDNA fraction. One patient did not have ctDNA results (black box). The ctDNA fraction is shown in red/green, and the CTC count displayed below in black. Blue boxes represent tumors that were sequenced. Patient 2 had basal cell skin cancer resected. Patient 3 had seminoma treated with orchiectomy and radiation at age 22 yr. Patient 4 had squamous cell skin cancer resected. Patients 8 and 13 had Gleason 6 prostate cancer (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). CHEMO = chemotherapy; EBRT = external beam radiation therapy; LNU = left nephroureterectomy; MET = metastatic; NAC = neoadjuvant chemotherapy; NephroU = nephroureterectomy; NMIBC = non–muscle-invasive bladder cancer; RC = radical cystectomy; TURBT = transurethral resection of bladder tumor.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 E U R O P E A N U RO L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
morphology have started to emerge [5]. For ctDNA, there is also no consensus on the optimal strategy for detection. A personalized medicine approach of unique droplet digital polymerase chain reaction (ddPCR) assays targeted to each individual patient’s known mutations should offer greater sensitivity than a generic next-generation sequencing (NGS) panel. Conversely, an NGS panel may be able to detect mutations that have arisen later and are not represented in the primary tumor. Compounding these considerations is the distinctly lower expression of CTCs and ctDNA in UC than in other malignancies [6]. Given the potential of these distinct liquid biopsies for risk stratification and the lack of information on their relative utility in UC, we performed a head-to-head comparison of a selection-free immunofluorescence-based CTC assay (RareCyte) and a plasma NGS panel for ctDNA detection in a pilot cohort of 16 consecutive patients with metastatic UC. Fig. 1 shows their clinical course and treatment. We found similar detection rates for CTCs (75% of patients; median 2.5 CTCs, range 0–170) and ctDNA (73% of patients; median of 2 somatic mutations, range 0–7) yet interestingly the CTC count did not
3
correlate with cell-free DNA or the ctDNA fraction (Supplementary Fig. 1). We discovered a similar genomic landscape to that previously described, with frequent mutations in TP53, TERT, and ERBB2 (Fig. 2) [7]. With a median follow up of 5.9 mo, a ctDNA fraction of >2% was associated with worse overall survival (OS; p = 0.039) with a median OS of 38 wk versus not reached for ctDNA 2% (Supplementary Fig. 2). By contrast, CTC detection did not have a significant impact on OS (p = 0.46), with a median OS of 59 wk for no CTCs versus 25 wk for 1 CTC (Supplementary Fig. 3). The cutoffs were defined a priori on the basis of previous work [7]. Notably, our ctDNA detection rate of 73% is in line with that of three other groups that reported 65%, 69%, and 73% for small cohorts [8]. Overall, these numbers are disappointing, as they represent incomplete detection in advanced disease, and the ideal liquid biopsy should be sensitive enough for early diagnosis and detection of minimal residual disease. Of note, there is one commercially available NGS ctDNA panel (Guardant360) with a detection rate of 90% reported for genomic alterations in advanced UC; however, the same group showed that the results for
Fig. 2 – Mutational landscape of metastatic bladder cancer plasma samples. Each column represents one plasma sample for a given patient, with the corresponding circulating tumor DNA fraction plotted on the bar graph above. Each mutated gene is shown on a row, with the mutation frequency of that gene in the overall cohort shown on the bar graph to the right. Boxes are color coded with respect to the mutant allele fraction of the mutation. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 4
E U R O P E A N U R O L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
this panel were significantly discordant with tumor collected at the same time [9–11]. In the present study we found higher mutant allele frequencies (MAFs) for plasma mutations that were concordant with tissue (11.16 vs 0.40; p 0001) and that were triplicated in replicates (43.38 vs 21.09; p = 0.07), suggesting a certain threshold for detection with the NGS assay. Logically, it might be expected that a tissue-personalized ddPCR assay would provide greater sensitivity; however, this approach also has limitations due to tumor heterogeneity and clonal evolution. In this context, in the present study several potentially clinically actionable mutations were only detected in plasma and not in matched tumor specimens collected at various time points relative to the liquid biopsies (91% of patients had tissue confirmation of at least one plasma mutation, Supplementary Figs. 4 and 5; all available primary/metastatic tumor was sequenced, Supplementary Fig. 1). The main example of a tissue-personalized ddPCR assay in UC was described by BirkenkampDemtröder et al [12]. Five of 12 patients (42%) had at least one undetectable result during metastatic disease, with one patient having seven negative reads. These patients received heterogeneous treatments, and the question of sensitivity for the metastatic setting of ddPCR assays designed based on primary tumor alterations deserves further dedicated study. An additional challenge for NGS assays is that quantification of mutant genome expression must be extrapolated from the MAFs. By design, MAF is affected by the overall coverage of the region of interest, which is not uniform across the genome. This limitation affected the present study, and it must be emphasized that our analysis was certainly impacted by small numbers and heterogeneous clinicopathologic factors at baseline. CTC detection was not significantly associated with OS; however, this could simply reflect a lack of statistical power necessary to detect a relationship that does exist. Despite these challenges, liquid biopsy represents the next frontier for diagnosis, management, and follow-up for solid malignancies. Our results suggest that CTCs and ctDNA can provide complementary information: ctDNA quantification was relevant for survival and the NGS assay detected individual patient mutations that were not identified in tumor, whereas the CTC platform demonstrated feasibility for capture of immunoflourescent images revealing the morphology of circulating malignant cells as well as protein expression at the single cell level. When considering the ideal CTC assay, the emphasis should shift from simple enumeration to a biological assay. It has been demonstrated in preclinical studies that the CTC count fluctuates from 0 to 54 in a 5-min time span [13], and in the present study we measured median fluctuations from moment to moment and day to day of 2 and 7 CTCs, respectively. Using the Epic CTC platform, Scher et al [5] showed that the degree of cell phenotype heterogeneity in prostate cancer can inform treatment response to taxane versus androgen inhibitor therapy, and similar studies in UC are certainly warranted. Finally, we have shown that quantitative measures of ctDNA may be used for patient risk stratification, although qualitative mutation results can also have important implications for therapy decisions. We have
described the unique challenge of ctDNA detection in UC and noted that the optimal approach must provide sensitivity without losing the ability to detect clonal evolution. Author contributions: Heather J. Chalfin had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Chalfin, Glavaris, Gorin, Hahn, McConkey. Acquisition of data: Chalfin, Glavaris, Kates, Fong, Dong, Matoso, Bivalacqua, Johnson, Pienta, Hahn, McConkey. Analysis and interpretation of data: Chalfin, Glavaris, Gorin, Dong, Hahn, McConkey. Drafting of the manuscript: Chalfin, Glavaris. Critical revision of the manuscript for important intellectual content: Chalfin, Glavaris, Gorin, Kates, Bivalacqua, Pienta, Hahn, McConkey. Statistical analysis: Chalfin, Glavaris, Gorin. Obtaining funding: Chalfin, Hahn, McConkey. Administrative, technical, or material support: Glavaris, Fong, Dong, Matoso. Supervision: Gorin, Kates, Bivalacqua, Johnson, Pienta, Hahn, McConkey. Other: None. Financial disclosures: Heather J. Chalfin 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: Kenneth J. Pienta has acted as a consultant for Celsee Diagnostics and acts as the Chief Medical Officer for Cuebiophrma. The PlasmaSELECT test is owned by Personal Genome Diagnostics (PGDx). The Johns Hopkins University has licensed technologies to PGDx and, as a result, has a financial interest in the company. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies. The remaining authors have nothing to disclose. Funding/Support and role of the sponsor: H.J.C. is supported by a grant from the Johns Hopkins Greenberg Bladder Cancer Institute. The sponsor played a role in review of the manuscript.
Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j. euo.2019.08.004.
References [1] Howlader N, Noone AM, Krapcho M, editors. SEER cancer statistics review, 1975–2013. Bethesda, MD: National Cancer Institute; 2016. http://seer.cancer.gov/csr/1975_2013/ [2] deVere White RW, Lara Jr PN, Goldman B, et al. A sequential treatment approach tomyoinvasive urothelial cancer: a phase II Southwest Oncology Group trial (S0219). J Urol 2009;181:2476–80. [3] Gallagher DJ, Milowsky MI, Ishill N, et al. Detection of circulating tumor cells in patients with urothelial cancer. Ann Oncol 2009;20:305–8. [4] McDaniel AS, Ferraldeschi R, Krupa R, et al. Phenotypic diversity of circulating tumour cells in patients with metastatic castrationresistant prostate cancer. BJU Int 2017;120:E30–44.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004
EUO-262; No. of Pages 5 E U R O P E A N U RO L O GY O N C O L O GY X X X ( 2 019 ) X X X – X X X
5
[5] Scher HI, Graf RP, Schreiber NA, et al. Phenotypic heterogeneity of
[10] Barata PC, Koshin VS, Funchain P, et al. Next-generation sequencing
circulating tumor cells informs clinical decisions between AR sig-
(NGS) of cell-free circulating tumor DNA and tumor tissue in
naling inhibitors and taxanes in metastatic prostate cancer. Cancer
patients with advanced urothelial cancer: a pilot assessment of
Res 2017;77:5687–98. [6] Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating
concordance. Ann Oncol 2017;28:2458–63. [11] Grivas P., Lalani A.A., Pond G.R., et al. Circulating tumor DNA
tumor DNA in early- and late-stage human malignancies. Sci Transl
alterations in advanced urothelial carcinoma and association with
Med 2014;6:224ra24.
clinical outcomes: a pilot study. Eur Urol Oncol. In press. http://dx.
[7] Vandekerkhove G, Todenhöfer T, Annala M, et al. Circulating tumor DNA reveals clinically actionable somatic genome of metastatic bladder cancer. Clin Cancer Res 2017;23:6487–97. [8] Tan MP, Attard G, Huddart RA. Circulating tumour DNA in muscleinvasive bladder cancer. Int J Mol Sci 2018;19:2568. [9] Agarwal N, Pal SK, Hahn AW, et al. Characterization of metastatic
doi.org/10.1016/j.euo.2019.02.004. [12] Birkenkamp-Demtröder K, Christensen E, Nordentoft I, et al. Monitoring treatment response and metastatic relapse in advanced bladder cancer by liquid biopsy analysis. Eur Urol 2018;73:535–40. [13] Fei F, Du Y, Di G, Wu J, Shao Z. Are changes in circulating tumor cell (CTC) count associated with the response to neoadjuvant chemo-
urothelial carcinoma via comprehensive genomic profiling of cir-
therapy in local advanced breast cancer? A meta-analysis. Oncol Res
culating tumor DNA. Cancer 2018;124:2115–24.
Treat 2014;37:250–4.
Please cite this article in press as: Chalfin HJ, et al. Circulating Tumor Cell and Circulating Tumor DNA Assays Reveal Complementary Information for Patients with Metastatic Urothelial Cancer. Eur Urol Oncol (2019), https://doi.org/10.1016/j. euo.2019.08.004