Frequency and Prognostic Value of Circulating Tumor Cells in Cancer of Unknown Primary

Abstract

BACKGROUND

Cancer of unknown primary (CUP) is defined as a primary metastatic malignancy, in which the primary tumor remains elusive in spite of a comprehensive diagnostic workup. The frequency and prognostic value of circulating tumor cells (CTCs), which are considered to be the source of metastasis, has not yet been systematically evaluated in CUP.

METHODS

A total of 110 patients with a confirmed diagnosis of CUP according to the European Society for Medical Oncology (ESMO) guidelines, who presented to our clinic between July 2021 and May 2023, provided blood samples for CTC quantification using CellSearch methodology. CTC counts were correlated with demographic, clinical, and molecular data generated by comprehensive genomic profiling of tumor tissue.

RESULTS

CTCs were detected in 26% of all patients at initial presentation to our department. The highest CTC frequency was observed among patients with unfavorable CUP (35.5%), while patients with single-site/oligometastatic CUP harbored the lowest CTC frequency (11.4%). No statistically significant association between CTC positivity and the number of affected organs (P = 0.478) or disease burden (P = 0.120) was found. High CTC levels (≥5 CTCs/7.5 mL; 12/95 analyzed patients) predicted for adverse overall survival compared to negative or low CTC counts (6-months overall survival rate 90% vs 32%, log-rank P < 0.001; HR 5.43; 95% CI 2.23–13.2). CTC dynamics were also prognostic for overall survival by landmark analysis (log-rank P < 0.001, HR 10.2, 95% CI 1.95–52.9).

CONCLUSIONS

CTC frequency is a strong, independent predictor of survival in patients with CUP. CTC quantification provides a useful prognostic tool in the management of these patients.

Introduction

Cancer of unknown primary (CUP) represents a heterogeneous group of malignancies in which metastatic spread is histologically proven; the site of origin, however, remains undetectable despite extensive diagnostic workup (1). Although accounting for only 3%–5% of all malignancies, CUP constitutes a major diagnostic and therapeutic challenge. The clinical presentation can vary widely and is generally characterized by early metastatic dissemination in an unusual pattern, an aggressive clinical course, and resistance to empiric chemotherapy. Based on clinical and histological criteria, CUP can be classified into a favorable and an unfavorable subgroup (1). Favorable subtypes either exhibit obvious analogies to metastatic malignancies with known primaries and can therefore achieve prolonged survival following treatment according to the corresponding cancer type, or are oligometastatic and amenable to local ablative treatment. Accordingly, favorable CUP subtypes are classified into breast-like, ovary-like, head and neck-like, prostate-like, colon-like, renal-like and single-site/oligometastatic CUP. The much larger subgroup of unfavorable CUP, which makes up 80%–85% of cases, is defined by the absence of criteria allowing categorization into one of the favorable subtypes, typically demonstrates extensive metastatic dissemination, and is usually treated with empiric platinum-based chemotherapy. As a result, unfavorable CUP is associated with a median overall survival (OS) of only 3–9 months and a 5-year survival rate of <10% (1).

Given its elusive nature, understanding the biology of CUP and the mechanisms leading to early metastatic seeding in the absence of a recognizable primary tumor has proven a demanding task. Several therapeutic approaches have been proposed, mainly with the goal to identify the primary tumor to allow for tissue-of-origin-specific treatment. However, recent randomized clinical trials failed to demonstrate a survival benefit of primary site-specific therapy following gene expression profiling-based tissue-of-origin prediction (2, 3). Current trials are instead turning to a primary site-agnostic approach, investigating the role of molecularly targeted therapies based on the fact that about 30% of CUP tissues harbor at least one actionable alteration (4, 5). This strategy might therefore improve prognosis for a relevant subpopulation of patients with unfavorable CUP. However, since such alterations are not universally present, there is still an urgent need to better understand CUP pathophysiology to further improve the management options for this still devastating disorder.

Over the past decades, the detection and characterization of circulating tumor cells (CTCs) in the blood of cancer patients have emerged as a promising approach to cancer monitoring and prognostication (6). Since metastatic spread typically occurs as a result of direct intravasation of tumor cells into the blood stream or the lymphatic system, CTCs may represent disseminating cells, being responsible for metastatic spread.

The detection of CTCs has been associated with poor prognosis and decreased OS in various malignancies including breast, lung, prostate, and colorectal cancer (7–10). In addition, CTCs have been shown to be a valuable tool for monitoring response to treatment, predicting the risk of relapse, and guiding personalized treatment decisions (11). By contrast, although metastasis formation is central to the pathogenesis of CUP, frequency and prognostic significance of CTCs in CUP have never been systematically analyzed. Only a few studies on a handful of patients have reported the detection of CTCs in CUP patients up to now (12, 13).

In this prospective study, we investigated the frequency and prognostic value of CTCs in 180 blood samples from a cohort of 110 CUP patients. CTC frequency at diagnosis and during the course of treatment was determined and correlated with clinical and molecular data. Our findings suggest that CTCs are detectable in a significant proportion of CUP patients and that high CTC counts are associated with a poor prognosis. Our study highlights the potential of CTCs as a noninvasive diagnostic and prognostic tool in CUP, paving the way for further research in this field.

Materials and Methods

Patients

Patients with a confirmed diagnosis of CUP according to the European Society for Medical Oncology (ESMO) clinical practice guidelines, who presented to our clinic between July 2021 and May 2023 were eligible. Both, patients with newly diagnosed CUP and patients that presented to our department during or after treatment were included. The total cohort of 110 patients consists of 62 and 48 patients with unfavorable and favorable CUP, and 44 and 66 patients with newly diagnosed and pretreated CUP, respectively (Table 1). Favorable subtypes include 35 patients with single-site/oligometastatic disease and 13 patients with obvious analogy to certain primary tumor entities. In cases in which a relapsed antecedent malignancy could possibly have been mistaken for CUP, comparative panel sequencing of both tumors was performed as previously described to exclude relapses of prior malignancies (14). Serial blood samples were collected at first diagnosis of CUP, before starting a new therapy line, at every response assessment, and every 3–6 months following end of treatment. For patients treated with surgery or radiotherapy in curative intention, blood samples were collected directly before ablative treatment, 4–8 weeks following end of treatment, and every 3–6 months thereafter in parallel to follow-up imaging. CTC counts were correlated to predefined clinical characteristics, survival, and molecular tumor profiles. To better reflect the disease burden of the analyzed patients, we have developed a metastasis burden score as previously described, which includes the diameters of the largest lesion and the numbers of affected organs and metastases per organ, as these parameters have been shown to affect CUP patient prognosis (15–17). Metastasis burden score calculation and patient categorization criteria are detailed in Supplemental Table 1 in the online Data Supplement. All patients provided written informed consent. The study was conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki.

Isolation and Enumeration of CTCs, Panel Sequencing of Biopsy Samples, and Statistical Analysis

Cell isolation, enumeration, and sequencing methods as well as data analysis are presented in the Supplemental Material.

Results

Patient Demographics

A total of 110 patients with a confirmed diagnosis of CUP according to the current ESMO guidelines, who presented to our clinic between July 2021 and May 2023 were enrolled (1). Demographic and clinical characteristics of the study cohort are displayed in Table 1. Sixty-two and 48 patients were diagnosed with unfavorable (56.4%) and favorable (43.6%) CUP, respectively, with single-site/oligometastatic CUP (35 patients, 31.8%) representing the most prevalent favorable subtype in the analyzed cohort. Disease histology mainly included adenocarcinoma (70 patients, 63.6%), squamous cell carcinoma (26 patients, 23.6%) and undifferentiated carcinoma (7 patients, 6.4%). EpCAM (epithelial cell adhesion molecule) expression of tumor cells was assessed in 10 cases, 7 of which were at least partially positive. Only one of these patients was CTC^pos. The organs most commonly affected by metastases were lymph nodes (50% of patients), pleura or peritoneum (30.9%), lungs (22.7%), and liver (27.3%). Forty-four of 110 baseline samples, defined as the first blood samples taken for CTC analysis of patients included into this study, were collected at first diagnosis of CUP, 23/110 at disease progression, 26/110 during treatment, and 17 samples after completion of a therapy line. The most common treatment strategies in the overall cohort were palliative chemotherapy (55 cases), and ablative surgery followed by adjuvant radiotherapy (8 cases) or ablative surgery alone (7 cases) in patients with single-site/oligometastatic and head/neck-like CUP.

Survival analysis included 95 patients who presented in our department between July 2021 and April 2023. Demographic and clinical characteristics of these patients are summarized in Supplemental Table 2. At the time of data cutoff (April 2023) 25 of these patients had died. The 6-months and 1-year OS rates, as calculated from the time of baseline blood sampling, were 84% (95% CI 76%–92%) and 62% (95% CI 51%–76%), respectively, after a median follow-up of 8 months (range 0–20.1 months). Patients suffering from unfavorable CUP had a 6-months OS rate of 72% (95% CI 59%–87%) compared to 97% (95% CI 91%–100%) for patients with favorable CUP subtypes (log-rank P < 0.001; HR 4.57; 95% CI 1.71–12.2). Disease burden was significantly associated with OS, as patients with high metastasis burden scores reached a 6-months OS rate of only 71% (95% CI 56%–90%) compared to 79% (95% CI 64%–97%) for patients with intermediate disease burden, while the 6-months OS rate was 100% (95% CI 100%–100%) for the group of patients with low metastasis burden scores (log-rank P < 0.001, HR intermediate metastasis burden score: 3.25, 95% CI 0.84–12.6; high metastasis burden score: 9.01, 95% CI 2.58–31.5; Supplemental Fig. 1A).

Detection of CTCs in Patients with CUP and Association with Clinical Characteristics

Patients provided 180 blood samples including 110 baseline and 70 follow-up samples. Twenty-three patients provided one and 17 patients multiple follow-up samples (2 follow-up samples: n = 9; >2 follow-up samples, n = 8). CTCs were detected in 36 patients (32.7%) and 43 samples (23.8%) of the overall cohort. Baseline CTCs were present in 26.4% of patients (29/110). Histology had a substantial impact on the rate of CTC positivity. While 35.7% of patients with adenocarcinomas were CTC^pos, squamous cell histology was associated with CTC positivity in only 7.7% of cases (P = 0.010). Patients with an intermediate or high metastasis burden score demonstrated positive CTC counts in 23 of 71 cases (32.4%) compared to 6 of 39 cases with low metastasis burden score (15.4%, P = 0.120). However, the effect of disease burden on CTC positivity did not reach statistical significance. Organ types affected were not significantly associated with CTC positivity.

Patients suffering from unfavorable CUP were CTC^pos at baseline in 22/62 (35.5%) cases with CTC counts varying between 1 and 585 CTCs/7.5 mL (median 6 CTCs/7.5 mL). Favorable subtypes were associated with a significantly lower rate of CTC positivity (7/48 patients, 14.6%, P = 0.007) and lower CTC counts in CTC^pos patients (median 1, range 1–17 CTCs/7.5 mL, P = 0.034).

Among 35 patients with single-site/oligometastatic CUP, baseline CTCs were detected in only 4 cases (11.4%), with counts ranging between 1 and 2 CTCs/7.5 mL (median 1 CTCs/7.5 mL; Table 1). In all 4 patients, the baseline CTC^pos blood sample was harvested after ablation of the single-site/oligometastatic CUP lesion. Within this subgroup, CTC^pos patients demonstrated neither a higher disease burden nor longer time periods between ablative treatment and blood sampling for CTC detection as compared to CTC^neg patients. As of data cutoff, none of the 4 CTC^pos patients had experienced disease progression or relapse. In contrast, all 14 patients with single-site/oligometastatic CUP in whom the baseline blood samples were harvested before ablative treatment were CTC^neg. Among 31 CTC^neg patients, 3 progression-free survival (PFS) events were documented. No CTCs were detected in the 4 patients with head and neck CUP. Both patients diagnosed with ovary-like CUP were CTC^pos (100%), while one of 7 patients with colon-like CUP had positive CTC counts (14.3%).

Baseline CTC Counts and Prognosis

Among the 95 patients included in the survival analysis, 9 CTC^pos and 16 CTC^neg patients died during follow-up. Patients with CTCs detected at baseline demonstrated a 6-months OS rate of 67% (95% CI 49%–91%) compared to 90% (95% CI 82%–98%) for CTC^neg patients (log-rank P = 0.1, HR 1.88; 95% CI 0.83–4.26; Fig. 1A). Patients with low CTC counts (<5 CTCs/7.5 mL, CTC^low, 14/95 patients) had a favorable prognosis similar to CTC^neg patients, indicating that, similar to other cancer entities, low CTC counts do not significantly impair survival in CUP [6-months OS rate 90% vs 92% vs 32%; log-rank P < 0.001; HR CTC^neg vs CTC^low 0.59 (95% CI 0.14–2.59), CTC^neg vs CTC^high 5.05 (95% CI 2.05–12.5); Fig. 1B and C) (9, 18). Accordingly, upon comparing patients with negative or low CTC counts (0–5 CTCs/7.5 mL, CTC^neg/low) to patients with high CTC counts (≥5 CTCs/7.5 mL, CTC^high), CTC^neg/low patients demonstrated a statistically significant OS benefit [6-months OS rate 90% (95% CI 83%–97%) vs 32% (95% CI 11%–91%), log-rank P < 0.001; HR 5.43; 95% CI 2.23–13.2; Fig. 1D).

When considering the unfavorable CUP subgroup only, CTC^high patients demonstrated a significantly inferior OS compared to CTC^neg/low patients as well [6-months OS rate 42% (95% CI 18%–94%) vs 76% (95% CI 66%–94%), log-rank P = 0.03, HR 2.70, 95% CI 1.03–7.08; Fig. 2A)]. Within the favorable subgroup only one patient had CTC^high and demonstrated an OS of 5.7 months. In the remaining patients (CTC^neg/low) of this subgroup, only 4 OS events occurred and median OS was not reached after a median follow-up of 9.2 months. Compared to unfavorable subtype/CTC^high and unfavorable subtype/CTC^neg/low patients, favorable subtype/CTC^neg/low patients showed a significantly increased OS (Fig. 2A). Median OS and 1-year OS rates were 4.8 months (95% CI 2.3-NA) and 21% (95% CI 4.2%–100%) for unfavorable subtype/CTC^high cases, not reached (95% CI 8.7-NA) and 50% (95% CI 34%–74%) for unfavorable subtype/CTC^neg/low cases, and not reached and 86% (95% CI 74%–100%) for favorable subtype/CTC^neg/low cases, respectively.

In 39/95 cases, CTC samples were collected and analyzed at first diagnosis of CUP. Thirty-three of these patients were CTC^neg/low and 6 patients were CTC^high. High CTC counts at first diagnosis were associated with a significantly reduced OS [6-months OS rate 33% (95% CI 11%–100%) vs 96% (95% CI 88%–100%), log-rank P < 0.001, HR 13.1, 95% CI 3.25–52.9; Supplemental Fig. 1B].

Fifty-four patients were evaluable for PFS analysis, since they provided blood samples directly before the initiation of a new therapy line. Treatment modalities in this group included chemotherapy in 31 cases (57.4%), immunotherapy in 4 cases (7.4%), and targeted therapies in 4 cases (7.4%). Thirty PFS events were documented. The overall median PFS was 6.5 months (95% CI 4.9-NA). CTC status affected PFS, as CTC^high patients demonstrated a median PFS of only 3.6 months (95% CI 2.5-NA) compared to 7.4 months (95% CI 5.3-NA) for CTC^neg/low patients. Likely due to the low number of CTC^high patients, this effect did not reach statistical significance (log-rank P = 0.1, HR 1.98, 95% CI 0.79–4.92; Fig. 2B).

CTC Dynamics and Survival

Forty patients provided follow-up CTC samples. Among these, 23 samples were collected up to 4 months after initial sampling and were therefore available for landmark analysis at the 4-months landmark timepoint. Patients were categorized into groups with negative or decreasing CTCs (CTC^dec, positive to negative or negative to negative) and positive or increasing CTCs (CTC^inc, negative to positive or positive to constantly or increasingly positive) depending on their CTC dynamics between first and second sample collection. CTC dynamics were highly prognostic for OS, as CTC^dec patients achieved a statistically significantly higher 6-months OS rate compared to patients in the CTC^inc group [93% (95% CI 82%–100%) vs 33% (95% CI 11%–100%), log-rank P < 0.001, HR 10.2, 95% CI 1.95–52.9; Fig. 3).

Association of CTC Detection with Molecular Characteristics

Panel sequencing data were available for 76/95 patients (80%). In 63 cases, TSO500 panel sequencing was performed. Five tumor samples were analyzed using the FoundationOne Tissue panel and in one case whole-exome sequencing data were available. In the remaining 9 cases, smaller panels were used due to scarce tissue material. The most common molecular alterations involved the TP53 (26 of 75 cases, 34.4%), CDKN2A (20 of 74 cases, 27.0%), PIK3CA (14 of 74 cases, 18.9%), KRAS (15 of 76 cases, 19.7%), KMT2D (9 of 68 cases, 13.2%), and KMT2C genes (5 of 68 cases, 7.4%). No significant differences in the prevalence of these alterations were observed between CTC^pos and CTC^neg patients (Table 2).

Multivariate Analysis of Risk Factors on Overall Survival

In multivariate analysis, the effect of high CTC counts on OS was only slightly reduced but sustained statistical significance after adjusting for disease histology and prognostic subgroup (HR 4.11, 95% CI 1.56–10.84, P = 0.004; Fig. 4). Unfavorable CUP remained a strong predictor of inferior survival (HR 4.01, 95% CI 1.43–11.24, P = 0.008).

Discussion

This study presents a comprehensive analysis of the frequency, association with clinical factors, and prognostic relevance of CTCs in CUP. Our results indicate that, overall, CTCs are present in 34% of patients with CUP and constitute a clinically relevant, independent risk factor for inferior survival.

Metastatic disease accounts for most cancer-related deaths (19). As a central part of the metastatic cascade, a fraction of CTCs survives within the circulation, manage to extravasate at distant sites, and are therefore considered to be the drivers of metastasis (20). Although CUP represents the paradigm metastatic disorder and despite extensive CTC research in other cancer entities, CTCs have not yet been systematically evaluated in CUP.

In our cohort of 110 patients, CTCs were present in first available blood samples of 26.4% from all patients and 35.5% from patients with unfavorable CUP, which is slightly lower compared to previously published data on smaller CUP cohorts reporting 50% CTC positivity using the EpCAM-dependent CellSearch technology in a cohort consisting of 5 untreated and 5 pretreated patients with unfavorable CUP and 62% in 24 patients that were sampled at random timepoints during disease course and analyzed by conventional immunofluorescence microscopy (12, 13). Although CUP represents a primary metastatic malignancy, the frequency of CTC detection was lower than described in other cancer entities, in which EpCAM-based detection methods typically report CTC positivity in about 48%–69% of patients with advanced lung, colorectal, breast, or prostate cancer (7–9, 21). In our study, the EpCAM expression status of the tumor tissue was assessed in only 10 cases as part of the routine patient care. Out of 10 cases, 7 showed EpCAM positivity, of which only one case was CTC^pos. Another factor contributing to the comparatively low rate of CTC positivity might be the relatively high fraction of patients with single-site/oligometastatic CUP, of whom in 21/35 (60%) cases baseline blood samples were harvested only after ablation of the single-site/oligometastatic CUP lesion.

In our cohort, the effect of disease burden on CTC positivity did not reach statistical significance in neither the entire cohort nor the unfavorable subgroup. By contrast, CUP prognostic subgroups were strongly associated with CTC levels. Accordingly, favorable subgroups demonstrated a lower CTC positivity rate as well as lower CTC counts in CTC-positive cases compared to unfavorable CUP. Moreover, single-site/oligometastatic CUP, which was the most prevalent favorable subtype in the analyzed population, demonstrated the lowest rate of CTC detection (11.4%), with no correlation between CTC positivity and disease burden within this subgroup either. Of note, clinical parameters including tumor volume do not affect survival in single-site/oligometastatic CUP (22). Upon analyzing the effect of common genetic alterations on the incidence of CTC positivity, both CDKN2A and KRAS alterations, which are associated with adverse prognosis in CUP (16), were slightly more common in CTC^neg patients. These findings suggest that differences in disease biology of favorable and unfavorable CUP subtypes rather than disease burden or molecular profiles dictate CTC levels and prognosis in CUP (23).

Importantly, and similar to other cancer entities, CTC levels in CTC^pos patients appeared to be prognostically relevant: high CTC counts led to a significantly reduced OS, whereas low counts (1–5 CTCs/7.5 mL) did not affect survival probability, indicating that a cutoff of 5 cells per 7.5 mL blood may be applied to evaluate patient prognosis in unfavorable CUP. This finding therefore adds to the growing body of evidence that CTC levels predict survival probability across several cancer types including CUP (7–9, 21).

Among the patients with available follow-up samples, CTC dynamics were also predictive of OS, as patients that remained CTC negative or achieved CTC negativity during treatment had a significantly superior OS compared to patients with consistently positive or increasing CTC counts. As this is the first study to evaluate CTC dynamics in CUP, our results indicate that CTC counts may be informative in the management of CUP, in which prognostic information is often limited. Monitoring CTC levels over time may help to assess treatment response and guide clinical decision-making.

Our study is the first to provide evidence for the prognostic value of CTCs in a large cohort of CUP patients. CTC detection and monitoring may be a valuable tool for stratifying patient risk and tailoring treatment strategies to individual patients. To further determine the usefulness of CTC detection in the management of patients with CUP, future studies should additionally investigate EpCAM-independent CTC detection strategies, including the functional and prognostic relevance of detecting CTCs that have undergone epithelial-to-mesenchymal transition, by size-based CTC isolation systems (23, 24). Beyond quantification, the molecular characterization of CTCs might shed light on the underlying biology of CUP and guide treatment decisions. In addition to CTCs, first results indicate that both baseline and longitudinal on-treatment circulating tumor DNA analyses are of prognostic and predictive value in CUP (17, 25, 26), similar to other cancer entities, and can therefore potentially be combined with CTC data to further increase its prognostic potential.

Supplemental Material

Supplemental material is available at Clinical Chemistry online.

Nonstandard Abbreviations

CUP, cancer of unknown primary; CTC, circulating tumor cells; ESMO, European Society for Medical Oncology; OS, overall survival; EpCAM, epithelial cell adhesion molecule; CTC^neg, circulating tumor cell-negative; CTC^pos, circulating tumor cell-positive; CTC^low, circulating tumor cell-low; CTC^high, circulating tumor cell-high; PFS, progression-free survival; CTC^dec, circulating tumor cell-decreasing; CTC^inc, circulating tumor cell-increasing.

Author Contributions

The corresponding author takes full responsibility that all authors on this publication have met the following required criteria of eligibility for authorship: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Nobody who qualifies for authorship has been omitted from the list.

Maria Pouyiourou (Formal analysis-Lead, Writing-original draft-Lead, Data curation-Equal, Conceptualization-Supporting, Investigation-Equal), Tilmann Bochtler (Conceptualization-Supporting, Writing-review & editing-Equal), Cornelia Coith (Methodology-Supporting), Harriet Wikman (Methodology-Supporting, Writing-review & editing-Equal), Bianca Kraft (Conceptualization-Supporting, Writing-review & editing-Equal), Thomas Hielscher (Formal analysis-Supporting, Writing-review & editing-Equal), Albrecht Stenzinger (Writing-review & editing-Equal), Sabine Riethdorf (Methodology-Lead, Conceptualization-Supporting, Data curation-Equal, Formal analysis-Supporting, Writing-review & editing-Equal, Supervision-Equal), Klaus Pantel (Methodology-Supporting, Conceptualization-Supporting, Data curation-Equal, Formal analysis-Supporting, Project administration-Supporting, Writing-review & editing-Equal, Supervision-Equal), Alwin Krämer (Conceptualization-Lead, Project administration-Lead, Data curation-Equal, Formal analysis-Supporting, Writing original-draft-Supporting, Investigation-Equal, Supervision-Equal).

Authors’ Disclosures or Potential Conflicts of Interest

Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest.

Employment or Leadership

K. Pantel, Associate Editor for Clinical Chemistry, Association for Diagnostics & Laboratory Medicine.

Consultant or Advisory Role

A. Stenzinger, Agilent, Aignostics, Amgen, Astra Zeneca, Bayer, BMS, Eli Lilly, Illumina, Incyte, Janssen, MSD, Novartis, Pfizer, Qlucore, Roche, Seattle Genetics, Takeda, and Thermo Fisher; K. Pantel, Menarini and Hummingbird; A. Krämer, Roche, and BMS.

Stock Ownership

None declared.

Honoraria

K. Pantel, Agena, Novartis, Illumina, Menarini, Abcam, and Sanofi; A. Krämer, Roche; T. Bochtler, Roche.

Research Funding

This study was funded by grants from the Deutsche Krebshilfe (Priority Program Translational Oncology Grant No. 70115167) to A. Krämer and K. Pantel, and the Clinician Scientist Program, Faculty of Medicine, University of Heidelberg to M. Pouyiourou. A. Stenzinger received funding from Bayer, BMS, Chugai, and Incyte not related to the scope of this study; A. Krämer received funding from BMS, Molecular Health, and the National Center for Tumor Diseases (NCT), Heidelberg, Germany, not related to the scope of this study.

Expert Testimony

None declared.

Patents

None declared.

Role of Sponsor

The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, preparation of manuscript, or final approval of manuscript.

References

Author notes