Review: Biological relevance of disseminated tumor cells in cancer patients


  • Sabine Riethdorf,

    1. Institute of Tumor Biology, Center of Experimental Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
    Search for more papers by this author
  • Harriet Wikman,

    1. Institute of Tumor Biology, Center of Experimental Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
    Search for more papers by this author
  • Klaus Pantel

    Corresponding author
    1. Institute of Tumor Biology, Center of Experimental Medicine, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
    • Institute of Tumor Biology, University Medical Center Hamburg-Eppendorf, Martinistrasse 52, 20246 Hamburg, Germany
    Search for more papers by this author
    • Fax: +49-40-42803 5374.


The prognosis of cancer patients is largely determined by the occurrence of distant metastases. In patients with primary tumors, this relapse is mainly due to clinically occult micrometastasis present in secondary organs at primary diagnosis but not detectable even with high resolution imaging procedures. Sensitive and specific immunocytochemical and molecular assays enable the detection and characterization of disseminated tumor cells (DTC) at the single cell level in bone marrow (BM) as the common homing site of DTC and circulating tumor cells (CTC) in peripheral blood. Because of the high variability of results in DTC and CTC detection, there is an urgent need for standardized methods. In this review, we will focus on BM and present currently available methods for the detection and characterization of DTC. Furthermore, we will discuss data on the biology of DTC and the clinical relevance of DTC detection. While the prognostic impact of DTC in BM has clearly been shown for primary breast cancer patients, less is known about the clinical relevance of DTC in patients with other carcinomas. Current findings suggest that DTC are capable to survive chemotherapy and persist in a dormant nonproliferating state over years. To what extent these DTC have stem cell properties is subject of ongoing investigations. Further characterization is required to understand the biology of DTC and to identify new targets for improved risk prevention and tailoring of therapy. Our review will focus on breast, colon, lung, and prostate cancer as the main tumor entities in Europe and the United States. © 2008 Wiley-Liss, Inc.

Solid tumors derived from epithelial tissues such as breast, prostate, lung or colorectal carcinomas are the most frequent forms of cancer in industrialized countries. Most solid tumor patients undergo a complete resection (R0) of their primary tumor but they still harbor a considerable risk to die from subsequent metastatic relapse caused by minimal residual disease (MRD) not being eliminated by primary surgery, radio- or chemotherapy.

By applying highly sensitive and specific immunocytochemical and molecular methods, it is now possible to detect disseminated tumor cells (DTC) in bone marrow and circulating tumor cells (CTC) in peripheral blood at the single cell level in the background of millions of normal cells.1 Among the distant organs, bone marrow (BM) plays the most prominent role as indicator organ for MRD, because BM is easily accessible by needle aspiration through the iliac crest. BM might be a common homing organ for DTC derived from various types of epithelial tumors. Thus far unknown environmental and internal factors can promote their recirculation from the BM niche into other distant organs such as liver or lungs where they can find better growth conditions.2 In the peripheral blood, CTC are detectable months to years after complete removal of the primary tumors, indicating that these cells might circulate between different metastatic sites.1

Our current review will focus on the relevance of DTC in BM with an emphasis on studies in patients with breast, colon, lung, and prostate cancer as the most frequent solid cancer entities. In particular, we will review the currently available methods for the detection and characterization of DTC, summarize the still limited knowledge about the specific biological properties of these cells, and discuss the clinical relevance of DTC detection with regard to an improved individualized management of cancer patients.


BM, bone marrow; CK, cytokeratin; CRC, colorectal cancer; CTC, circulating tumor cells; DTC, disseminated tumor cells; ICC, immunocytochemistry; MAGE, melanoma associated antigen; NSCLC, nonsmall cell lung cancer; PSA, prostate specific antigen; RT-PCR, reverse transcriptase polymerase chain reaction; SCLC, small cell lung cancer.

Detection of DTC

Due to the extremely low concentration of DTC (1 × 10−5 to 10−6) in BM, a direct detection of these cells is usually not feasible without a prior enrichment step. The most commonly used enrichment procedure is gradient centrifugation of BM through, e.g., Ficoll-Paque. With this procedure one can separate red blood cells and granulocytes from mononuclear cells and possible DTC based on their buoyant density. The separation of granulocytes is especially important in RT-PCR analyses, as, e.g., CK20 is also expressed by granulocytes.3, 4 For the isolation of CTC from peripheral blood, size-based enrichment has also been described, e.g., by membrane filter devices such as ISET (isolation by size of epithelial tumor cells5, 6) or MEMS (micro electro-mechanical system)-based microfilter approaches.7

Another enrichment approach used either alone or after gradient centrifugation is immunomagnetic bead separation which can be performed as (i) positive selection of DTC using epithelium-specific antibodies (e.g., anti EpCAM) or (ii) negative depletion of hematopoetic cells (e.g., by antibodies against the common leukocyte antigen CD45). The advantages and disadvantages of these enrichment methods have been discussed elsewhere.8, 9

Immunological detection of DTC

Current methods to screen BM aspirates for DTC after the enrichment can be classified into cytometric/immunological and molecular approaches.1, 10, 11

Among the cytometric methods immunocytochemistry (ICC) is the most widely used approach.2, 10 ICC has the advantage to facilitate characterization of both cell size and shape as well as the nucleus-plasma relation of each individual event. Thereby illegitimate expression of the protein of interest in BM cells can be excluded as far as possible. Because of the absence of tumor-specific target antigens, monoclonal antibodies against various epithelium-specific antigens such as cytoskeleton-associated cytokeratins, surface adhesion molecules, or growth factor receptors are applied for the detection of carcinoma cells.1, 10 These antibodies bind to the tumor cells which can then be visualized directly or indirectly by fluorescent dyes or colorimetrically.

The most commonly used antibodies, such as A45-B/B3, AE1/AE3, and KL-1 are directed against various intracellular cytokeratins (CK), which are intermediate filament proteins that are expressed in all epithelial cells but usually not in haematopoietic cells. Different CK are specific for specific epithelial cell types, e.g., CK20, frequently used for the detection of CTC/DTC in colorectal cancer (CRC) patients, is not suitable for CTC/DTC detection in lung cancer patients.12, 13

The detection of DTC in BM is still not part of the routine tumor staging in the clinical practice; however, emerging data anticipate a future role of DTC detection for risk stratification and therapeutic monitoring of breast cancer patients.14–16 Nevertheless, the detection rates of DTC in BM from nonmetastatic breast cancer and other carcinoma patients vary considerably (Tables I–IV). This might reflect the different sensitivity but also specificity of the numerous detection methods and antibodies used thus far. To overcome these problems and to provide a basis for future multicentric clinical trials, a consensus concept for the detection and enrichment of DTC in BM of thus far only breast cancer patients has been proposed.17 The use of pan-anti-cytokeratin antibodies A45-B/B3 or AE1/AE3 against a wide spectrum of cytokeratins was recommended as standard application, thereby ensuring detection of DTC also in cells that have downregulated the expression of individual cytokeratins, e.g., in the course of epithelial-mesenchymal transition.17–19 Criteria to evaluate morphology and staining results after automatic microscopic screening using sophisticated imaging approaches have also been defined to avoid false-positive and false-negative results.17, 20–22

Flow cytometry-based methods have been used for the detection of CTC/DTC. However, very few studies have been conducted on larger patient cohorts. These analyses are hampered by the lower sensitivity of flow cytometry as compared with other methods and have therefore mainly been described for analysis of BM or blood taken from advanced stage cancer patients with overt metastases.23, 24

Table I. Detection of DTC in Breast Cancer
No. of patientsStageBM aspirateEnrichment procedureDetection markers (antibody)No. of cells analyzedRange of positive cellsDetection rate cases %Detection rate controls (%)Correlation to clinical/pathol. variablesMean follow-up (months)Prognostic/predictive valueReference
  1. IC, iliac crest; ST, sternum; ICC, immunocytochemistry; Q-RT-PCR, quantitative reverse transcriptase polymerase chain reaction; MAM, mammaglobin; EMA, epithelial membrane antigen; CEA, carcinoembryonic antigen; PSE, prostate-specific Ets factor; PIP, prolactin-inducible protein; ER, estrogen receptor; PR, progesterone receptor; LN, lymph node; G, grading; TS, tumor size/stage; NA, information not available; NS, not significant; ND, not determined; DFS, disease-free survival; OS, overall survival; DM/LM, distant/lymph node metastasis; DMSF, distant metastasis-free survival; LRSF, local relapse-free survival; DDFS, Distant disease-free survival; BCSS, breast cancer-specific survival; RFS, relapse-free survival; TRD, tumor-related death.

 155T1–46.1 ml,IC, STFicollCK18 (CK2)NAMeanM0: 2.8/3*104;M1: 8.7/3*10418%0/75 (0%)DM(p < 0.001)NDND186
 49≥T1IC, STFicollcell-surfaceantigens (C26, T16),pan-CK (AE1)NANA37%NDNSmean 29,median 30(range 12–38)early recurrence(p < 0.04)187
 100T1–4, M04–6 ml, ICLymphoprepEMA,pan-CKNANA38%NDNSmedian 34(range 7–65)RFS (p = 0.0005),OS (p = 0.017)188
 727T1–410–12 ml,ICFicollmucin TAG12 (2E11)4–5 × 1061–9055% (LN+), 31% (LN0)0/21 (0%)TS (p < 0.001),G (p = 0.002),LN (p = 0.001)median 36DDFS (p < 0.001), OS (p < 0.001)189
 109T1–4ICFicollbreast-associatedantigens: MBr1,MBr8, MOV8,MOV16, MluClNANATotal: 31%;during surgery: 38%; 2–4 weeksafter surgery:17%NDNSmedian 36(range 15–62)NS190
 350T1–T416 ml ICLymphoprepEMANANA25%NDTS (p = 0.008), LN (p = 0.005), vascular invasion(p = 0.001)median 150(range 11–180)RFS (p < 0.001),OS (p < 0.001)191
 552T1–3ICFicollpan-CK(A45-B/B32 × 1061–1223(median 3)36%2/191 (1%)TS (p < 0.001), number of LNM.(p < 0.001),tumor type(p < 0.001),G (p = 0.017),DM (p < 0.001)median 38(range 10–70)DMSF (p < 0.001), OS (p < 0.001),192
 393≥T15–15 ml,IC, STFicollCK+EMANANA42%0/20 (0%)Number of DM(p < 0.001),bone metastases(p = 0.028),LN (p = 0.001)median 75TRD (p =0.01),DSF (p = 0.0003), OS (p = 0.022)193
 554pT1-2N0/1,M05–10 ml,ICFicollCK8, 18, 19(clone 5D3)NA1–300N0: 31%N1: 37%NDTS (p < 0.0001),G (p = 0.034),lymph vesselinvasion (p = 0.005),blood vessel invasion(p = 0.047),ER (p = 0.026),Ki-67 (p < 0.0001),uPA (p = 0.016), PAI-1 (p = 0.03)10 yearsDFS (p < 0.0001),OS (p < 0.0001)194
 920T1–T440 ml, ICLymphopreppan-CK(AE1/AE3)2 × 106NA13.4%5/40 (12.5%)TS (p = 0.013),LN (p < 0.0005), vascular invasion(p = 0.045),HER2 (p = 0.024)NDND195
 817T1–440 ml, ICFicollpan-CK(AE1/AE3)2 × 106NA13%4/98 (4%)TS (p = 0.011),LN (p < 0.001), ER and/orPR(p = 0.051)median 49(range 0.5–85)DDFS (p < 0.001), BCSS (p < 0.001)196
 114T1–43–5 ml,IC or STFicollpan-CK(A45-B/B3)3 × 1061 to >1000(mean: 70,median: 6)59%NDmenopausal status(p = 0.024),ER (p = 0.026),TS (p = 0.025)median 28OS (p = 0.0004),DSF (p = 0.012)197
 228pT1, 2,pN0–3,pM0)3–8 ml, ICFicollpan-CK(A45-B/B3)2 × 106NAPersistentDTC: 13%NDNSmean 49.8RFS (p = 0.0003), DFS (p < 0.0001), OS (p = 0.002) Persistent DTC: OS (p = 0.0008)16
 265≥T12–8 ml, ICFicollpan-CK(A45-B/B3)2 × 1061–1.500(median 2)26%NDNSmedian 60.5(range 7–255)OS (p = 0.03)198
 131T1–3, M0ICFicollICC: pan-CK(A45-B/B3)QPCR: CK193–6 × 106 eachICC: 1-12ICC: 31%Q-RT-PCR: 41%Combined: 51%NDNS1212 (48) months after surgery 35% (59%) of patients had positive ICC/Q-RT-PCR results54
 112pT1–410–20 mlBiocollpan-CK(A45-B/B3)2 × 106NA83% beforechemotherapy24% after surgeryand chemo/endocrine therapyNDDTCpersistence:G (p = 0.02)mean 12(range 5–30)ND199
 621T1–4, M0ST or ICFicollpan-CK(A45-B/B3)3 × 106NA15%NDNSmedian 56(range 1–100)OS (p = 0.02),DMSF (p = 0.006), LRFS(p = 0.0009)200
 46T1–45 ml ST and ICBuffy coatMUC5B, CK19, CEANDNDMUC5B:19%,CK19: 41%,CEA: 17%NDNSmedian 34(range 12–31)ND201
 55+ 4 DCIST1–410 ml ICLymphoprepCK19NDNDDCIS: 25%,T1: 43%T2–4: 60%1/10 (10%)NSat least 4 yearsDSF (p = 0.004)202
 148T1–4,M0, M+9 ml ICFicollCK19, MAMNDNDCK19, M0: 23%,M1: 47%,MAM, M0: 16%,M1: 38%MAM: 0/13(0%)PR (CK19:p = 0.028,MAM;p = 0.026)mean 786 daysOS (CK19:p = 0.0045,MAM;p = 0.025)203
 195T1–420 mlnoneCK191 × 107ND12%1/34 (3%)NSmedian 72(range 1–99)Systemic RSF(p = 0.01),Overall RSF(p = 0.005).204
 177 (RT-PCR), 83 (ICC)T1–3, M03–6 ml ICFicollICC: pan-CK(AE1/AE3)RT-PCR: MAM,CEA, PSE, PIPNDNDICC: 6% RT-PCRcomb.: 11% MAM: 4%,PIP: 2.8%,PSE: 2.8%,CEA: 1.1%n = 49 forthresholddefinitionNDNDND205
Table II. Detection of DTC in Colorectal Cancer
No. of patientsStageBM aspirateEnrichmentprocedureDetectionmarkers (antibody)No. of cells analyzedRange of positive cellsDetection rate cases %Detection rate controls (%)Correlation to clinical/pathol. variablesMean follow-up(months)Prognostic/ predictive valueReference
  • *

    CA19-9: Lewis blood group antigenes, 17-1A: membrane antigen, CD54-0: membrane antigen.

  • IC, iliac crest; ST, sternum; IMB, immunomagnetic bead separation; GCC, guanylylcyclase; CEA, carcinoembryonic antigen; LN, lymph node; NA, information not available; NS, not significant; ND, not determined; DFS, disease-free survival; OS, overall survival; RR, recurrance rate.

 57I–IV (18 M1)mean 6.1 ml IC or STFicollCK18 (CK2)3 × 104mean 2.8in M0, mean 8.7 in M121%0/75 (0%)NS M0 21% vrs M1 22%, S (p > 0.05), N0 9% vrs N1 35%NDND186
 82NANANACK18 (CK2)NANA27%0/75 (0%)NANDND206
 156Duke A-Dmean 5.2 ml ICFicollCK18 (CK2)1.5 × 105ND27%0/102 (0%)LN and stagemean 26RR (p < 0.05)207
 88I–IV, all R05ml ICFicollCK18 (CK2)3 × 105ND32%0/102 (0%)NSmedian 35 (range: 12–58)RR (p < 0.0035)150
 12 M1+ 7 post operativeI–IV5 ml ICFicollCK18 (CK2)1 × 1061–6016% M1, 71% post op.NDNDmedian 27 (range:11–32)NS208
 57I–IV0.5–10 ml IC or STFicollCK18 (CK2), pan-CK (A45-B/B3)4 × 105 to 1.6 × 106NDCK2: 16%,A45-B/B3: 43.5%CK2 and A45-B/B3: 4/75 (5.3%)NDNDND148
 67I–IV8 ml ICFicollCEA (C1P83), mucin (Ra96),pan-CK (KL-1), 17-1-A, CA19-9, CD54*each: 2.5 × 105ND29%0/25 (0%)NSNDND144
 48I–IIIICFlow cytometryCK181 × 105NDpreop.: 23% postop.: 27%1/63 (1.6%)NSNDRR (p < 0.01)for postoperative BM151
 34I–IV8 ml ICFicollpan-CK (KL-1),CK18 (CK2), CEA (C1P83),17-1-A*5 × 105NDtotal 74%CEA:30%, 17-1a:26, KL1:67%, CK2: 52%NDLN and stagemedian 12.5 (range: 6–18)NS145
 80all M1 (hepatic)NASmearA33, CK18, pan-CK(CAM 5.2, AE12)NANDresected: 9% nonres.: 34%0/20 (0%)NANDNS146
 109I–IV8 ml ICFicollCEA (C1P83), mucin (Ra96), pan-CK (KL-1), CA 19-9, 17-1A*each: 2.5 × 105ND49% (Ra96 and 17-1A and CEA:<10%, CA19-9:<15%, KI-1:<25%)KI-1, 17-1A, CEA: 2/45 (4.4%), CA 19-9: 1/45 (2.2%)stageNDNS147
 18Duke C and D4–10 ml ICFicollCK20 (Ks20.8) pan-CK (Cam 5.2)1 × 105CK 20: colon: 1.6*10–4,rectum: 5*10-5CK20:36% CK8: 0%CK18: 0%NDCK20 and CEA levelsin PTNDND209
 16784 N0, 43 M110 ml ICFicollpan-CK (A45-B/B3)1 × 1061–50024%0/51 (0%)NSmean 37 (range: 12–72)OS (p = 0.006), DFS (p < 0.001)152
 51I–IVICIMBpan-CK (A45-B/B3), EpCAM (Ber-Ep4), K-ras mutation2.5 × 106ND47%ND13/17 PT tumors with K-ras mutation also mutationin BMNDND67
 41Duke A-D1–3 ml ICLymphoprep + IMBpan-CK (A45-B/B3)1 × 1075–15 (mean 10)10%ND0% simultaneously stained with uPARNDND71
 275I–IV20 ml ICIMBEpCAM (MOC31)2 × 1072–120 (median 8)17%3/206 (1.5%)NSNDND149
 11Duke A-DICFlow cytometryCK18NAND38%NDNSNDND210
 47all M1 (hepatic)10 ml ICFicollpan-CK (A45-B/B3)3–4 × 1061–14 (median 1)55%NDNDmean 43 (range:26–54)NS153
 10NDNDnoneCEANDND66%0/56 (0%)NDNDND162
 15Duke A-DICnoneCK 20, CK 19NDNDCK19: 40% CK20: 0%CK19: 5/12 (42%),CK20: 0/12 (0%)NDNDND211
 57I–IV5 ml ICFicollCK 20NDND35%1/16 (6.2%)stageNDND159
 65I–IV, 40% M120 ml ICFicollCK 20NDND31%2/22 (9.1%)stageNDS in comb. with CTC in blood160
 14I–III10 ml ICFicollCK 20NDND21%NDNSNDND212
 30all M1 (hepatic)10 ml ICFicollCK 20NDND27%0/30 (0%)NDNDND213
 295I–IV, R0ICFicollCK 20NDND31%2/22 (9.1%)stageNAOS154
 109I–IV, 18% M12 ml ICnoneCK20, GCCNDNDCK20: 11%GCC: 6%CK20: 10/22 (45%), GCC:0/22 (0%)NSNDND156
 103 without +24 neoadjuv. therapyI–IV10 ml ICFicollCK 20NDND33% without and 17% with neoadjuvant treatmentNDNDmedian 49 (range:15–72)OS (p < 0.04) and DFS (p < 0.03) for neoadjuvant chemoradationpatients155
 32all M1 (hepatic)3–8 ml ICFicollCK 20NDND25%NDNSmedian 18 (range:5–40)NS157
 37all M1, R0 (hepatic)10 ml ICFicollCK 20NDND16%0/30 (0%)NDmedian 38 (range: 10–63)RR: 0.013, DFS (p = 0.04)161
 71all stage II, R010 ml ICFicollCK 20NDND28%0/30 (0%)NSmedian 58 (range:3–81)NS158
Table III. Detection of DTC in Lung Cancer
No. of patientsStageBM aspirateEnrichment procedureDetection markers (antibody)No. of cells analyzedRange of positive cellsDetectionrate cases %Detection rate controls (%)Correlation to clinicalvariablesMeanfollow-up (months)Prognostic relevanceReference
  • *

    Includes patients from ref. 168;

  • **

    Same study as ref. 163 but with longer follow-up;

  • ***

    Multicenter study with data collected from 15 hospitals.

  • IC, iliac crest; ST, sternum; IMB, immunomagnetic bead separation; GRP, prepro-gastrin-releasing peptide; NS, not significant; ND, not determined; DFS, disease-free survival; OS, overall survival; RR, recurrance rate.

 82I–III (N0 n = 42)0.5–5 ml ICFicollCK18 (CK2)2 × 106ND22%2/117 (1.7%)tumor size and grade13.0RR: (67% vs. 37%)168
 43I–IIIRibFicollpan-CK (CAM 5.2,AE1)1 × 106ND40%0/11 (0%)ND13.6DFS (p < 0.001),RR (p < 0.001)169
 139*I–III (N0 n = 70)2–10 ml IC, RibFicollCK18 (CK2)2 × 1061–531 (mean:2)all 60, N0:54%6/215 (2.8%)NS39.0RR: (p = 0.004) for N0 patients163
 39I–III (N0 n = 23)5 ml ICFicollCK18 (CK2)2 × 106ND38%0/5 (0%)NS4.6RR: (p = 0.008)170
 139**I–III (N0 n = 66)2–10 ml IC, RibFicollCK18 (CK2)2 × 1061–531 (mean:2)all 60%,N0:52%NDNS66.0OS (p = 0.007) for pN0 patients with ≥2CK2-positive cells164
 99I–IV (N0 n = 40, M1=3)8 ml ICFicollpan-CK (CAM 5.2,AE1)5 × 106ND22%NDNS14.3NS214
 58I–III (N0 n = 40)4 ml STFicollCK18 (CK2)1 × 106ND47%NDNS36.0OS (p = 0.044)165
 115I–II (all N0)5 ml ICFicollCK18 (CK2)1 × 106ND28%NDND35.8NS215
 80I–IV (n = n = 15, M1=27)10 ml ICFicoll+ IMBpan-CK (A45-B/B3)1–2 × 107ND23%NDND12.0OS (p = 0.030)171
 351***I–III5 ml ICFicollCK18 (CK2)1 × 1061–34 (mean:2)32%NDtumor size48.0OS (p = 0.047) in stage II–III172
 11 NSCLC, 7 SCLCI–IINDLymphopreppan-CK (CAM 5.2, MAK-6)0.5–1 × 105NDNSCLC: 0%, SCLC: 71%0/5, 0%NDNDNS216
 96I–IV (N0 n = 53, M1, n = 10)10 ml ICFicollpan-CK (AE1/AE3,MNF 116), EpCAM (Ber-Ep4)6 × 106ND22%1/32 (3.1%)NS50.5NS217
 196I–IV (M1, n = 91)10–20 ml ICLymphoprep+ IMBEpCAM (MOC31)2 × 1072–204 (mean:4)55%NDNS8NS173
 33T1–4 N0M01 mlnoneMAGE-A1-12NDND33%0/30 (0%)NDNDND63
 39 NSCLC, 32 SCLCI–IV2–3 mlFicollGRPNDNDNSCLC:0% SCLC: 18%0/28 (0%)NDNDND174
 50I–III, M0, R05 ml ICFicollMAGE-A1-12NDND52%0/30 (0%)grade 1 and 2 tumors92RR (p < 0.03) for N0 patients79
Table IV. Detection of DTC in Prostate Cancer
No. of patientsStageBM aspirateEnrichment procedureDetection markers (antibody)No. of cellsanalyzedRange ofpositive cellsDetection rate cases %Detection rate controls (%)Correlationto clinical/pathol. variablesMean follow-up (months)Prognostic/predictive valueReference
  1. IC, Iliac crest; S, significant; NS, not significant; ND, not determined; DFS, disease-free survival; DM, distant metastasis; OS, overall survival; RR, recurrance rate; RSF, recurrence-free survival; PSA, prostate specific antigen; GS, Gleason score; PE, prostatectomy; RT, radiotherapy; BS, bone scan; AD, androgen deprivation.

 84N0M06 ml ICFicollCK18 (CK2)8 × 105NA, mean 636%0/12 (0%)NANDLocal tumor extent, DM and differentiation61
 42NA0.5–10 ml ICFicollCK18 (CK2), pan-CK (A45-B/B3)2 × 106NACK2: 31%, A45-B/B3: 45.2%CK2: 4/75 (5.3%) A45-B/B3: 4/75 (5.3%)NANDND148
 44T3–4 N0M06–10 ml ICFicollCK18 (CK2)2 × 1061–38 (mean 6, median 2)55%NDNSNDND175
 36T3–4 M0 (34 N0, 2 N1, 2)3–10 ml ICFicollCK18 (CK2), pan-CK (A45-B/B3)2 × 1061–38 prior to therapy; 1–26 after therapy58% prior totherapy, 19 % after therapyNDNSNDND183
 48pT2–pT35 ml ICFicollpan-CK (CAM 5.2 + AE1/AE3 +NCL5D3)2.5 × 106NA10%0/20 (0%)NANDND218
 53T1–46–10 ml ICFicollCK18 (CK2)1.7 × 106NA38%NDNANDND62
 45T1–3 pN0M06–8 ml ICFicollCK18 (CK2)NANA20%NDNSMedian 25NS177
 154pT1–2 pN06–8 ml ICFicollCK18 (CK2)1–2 × 1061–5125.3%1/9 (11%)NAMedian 32NS141
 82N0–N16–8 ml ICFicollCK18 (CK2),pan-CK (A45-B/B3)1 × 106CK2: 1–21, A45: 1–128CK2: 20/82CK2 (24%), A45-B/B3 (40%)NDNSMedian 48Earlier biochemical relapse, p ≤0.004 for A45/B-B3180
 30T1–T3 N0M0ICFicollpan-CK (A45-B/B3)1 × 106NA30%NDNANDND63
 20pT2 pN0M0R0ICFicollpan-CK (A45-B/B3)NANA40%NDNANDND219
 220NA10 ml ICFicollEpCAM (BerEP4)NANA90% before PE, 69% 4 months after PE, 79% 5 years after PE1/48 (2%)NSNDND64
 55NA5–10 ml ICFicollpan-CK (A45-B/B3)2 × 106NA22%NDNANDND41
 266cT1–T4 pN0M020–30 ml ICLymphopreppan-CK (AE1/AE3)2 × 1061–1118–20%NDGS (p = 0.04), Gleason pattern (p = 0.04)Median 84Shorter RR (p = 0.03)in patients treated with RT, but no AD181
 292NA10 ml ICMACSEpCAM (BerEP4)NANA35–92%3/27 (11%)NS12NS65
 193cT1–3 pN0M05–8 ml ICFicollpan-CK (A45-B/B3)5 × 1061–50(median 2)45%NDNSMedian 16Shorter PSA-RSF(p = 0.003), predictive for PSA relapse (p = 0.014)179
 35T1–35 ml ICLymphoprepPSANDND27% (7/26 BS-), 56% (5/9 BS+)NDMetastatic bone disease (incl. micrometastasis) to serum PSA, BS- to serum PSA (p < 0.01)NDND220
 71pT1–3 M05 ml ICFicollPSANDND62%0/30 (0%)NSNDND218
 86cT1–23–5 ml ICFicollPSA1 × 107ND45%NDSerum PSA (p = 0.001), tumor stage (p = 0.003)Range: 1–43DSF (p = 0.004)176
 53ND6–10 ml ICFicollPSA3.2 × 106 28%0/53 (0%)NSNDND62
 116T1–28 ml ICFicollPSANDND44–66%NDNSNDRR182
 244cT1–25–10 ml ICFicollPSANDND32%NDNANDND90
 30T1–3 N0M01 mlnoneMAGE-A1-12, PSANDND60% (at least one MAGE-A-mRNA) 27% (PSA)0/30 (0%) (MAGE-A)High risk ofmetastatic relapse (p = 0.02)NDND63
 161pT1–41 ml ICnoneMAGE-A1-12NDND16%0/13 (0)%NSNDND78
 292ND10 ml ICMACSPSANDND43–83%3.7%NSNDNS65

Discrimination between viable and apoptotic DTC is facilitated by a new technique, designated EPISPOT (epithelial immunospot). This method offers the advantage to detect viable DTC by their ability to secrete individual proteins after short term culture (48 hr). In a recently published study, Muc-1 and/or CK19 secreting DTC could be demonstrated in BM samples of breast cancer patients with (90%) and without (50%) overt metastases but not from healthy controls.25 This technique has been also applied to CTC detection and its clinical validation is ongoing.

A series of new techniques has been developed and applied for CTC detection in the peripheral blood but not for BM analysis yet. Further progress toward a standardized method for CTC detection was reached through the introduction of the CellSearch system (Veridex, Warren, NJ), an automated enrichment and immunostaining device that has been cleared by the U.S. Food and Drug Administration for the detection of CTC in the peripheral blood of patients with metastatic breast, colon and prostate cancer.26–33 Recently, Nagrath et al.34 presented a microfluid platform called the “CTC-chip,” which consists of an array of anti-EpCAM antibody-coated microposts capable to capture CTC from unfractionated blood under precisely controlled laminar flow conditions.34 With this technique in more than 95% of cancer patients CTC were detected. This high incidence of CTC and the high number of CTC especially in patients without overt metastases requires further investigations on the specificity of this assay. A similar micropost-based method has been used commercially with rather limited success over several years by another group for the detection of fetal erythroblasts in the peripheral blood of pregnant women, another rare event application []. Furthermore, ultraspeed automated digital microscopy fiber-optic array scanning technology (FAST) and laser printing techniques have been developed to scan 300,000 cells per second, thereby detecting cells decorated by fluorescence dye-conjugated antibodies directly on a slide. However, additional time has to be assessed for artifact rejection and visual verification of the results.35–37 CTC counts 2–3 log units higher than other groups reported were detected by Pachmann et al.38, 39 in nearly 100% of breast cancer patients using the MAINTRAC assay. These discrepancies to the results obtained with other assays have raised discussions about the specificity of this approach using only anti-EpCAM antibodies for positive selection of tumor cells.38

Molecular detection of DTC

Molecular detection of DTC is based on PCR amplification of either DNA or cDNA (mRNA). A drawback of PCR-based methods is that the tumor cells of interest cannot be morphologically identified and isolated for further analyses.

DNA-based methods rely on the detection of known mutations, amplifications or methylation patterns in the tumor cells. Although numerous genetic alterations have been described, because of the tremendous heterogeneity of genetic alterations in tumor cells, currently no universally applicable DNA marker exists for DTC screening.1, 10, 11 Another possible confounder is the fact that DNA molecules are relatively stable, and therefore the detection of tumor specific genomic aberrations may not necessarily indicate the presence of intact DTC but also DNA fragments derived from necrotic or apoptotic tumor cells.40–42

Therefore, measurements of epithelium-specific or more organ-specific mRNA species such as CK19 and 20 or mammaglobin or PSA mRNA by RT (reverse transcriptase)-PCR have been proven as more promising to detect DTC in BM samples.43–50 The main advantage of very sensitive RT-based techniques for the detection of DTC at the single-cell level is the nearly “unlimited” availability of primers for almost every gene of interest. However, one of the major problems of RT-PCR approaches is low RNA stability. Therefore, storage and sample preparation have to be performed under conditions avoiding RNA degradation.50

RT-PCR based methods have been shown to be technically even more challenging than immunocytochemical assays with respect to specificity. The main drawback of using surrogate tissue-specific markers are false-positive results due to illegitimate low-level of epithelial- or tissue-related transcription in normal cells.51–53 Moreover, heterogeneity in the expression levels of a particular target transcript cannot be predicted. To avoid false-negative findings due to downregulation of the expression of a single gene, current RT-PCR analyses are therefore frequently performed as multimarker assays. Furthermore, quantitative real-time RT-PCR is helpful to decrease false positive findings by exclusion of very low DNA amounts deriving from illegitimate expression of noncancerous cells.45, 54–56 Marker genes currently used in RT-PCR approaches to detect DTC in BM from cancer patients are listed in several recent reviews.1, 8, 10

Specific biological characteristics of DTC

Biological characterization of cells categorized as DTC is directed to (1) confirm their malignant origin, (2) identify diagnostically and therapeutically relevant properties, and (3) increase the current knowledge about the biology of tumor cell dissemination in cancer patients. Since there is no gold standard thus far, different methods have been applied to assess the malignant potential of DTC. In addition to the technical challenges of single cell research, these studies are all hampered by the fact that these cells can exhibit features distinct from their corresponding primary tumors.

Previous reports indicated that DTC might be genomically heterogeneous.57 DTC can disseminate in a less-progressed genomic state and might aquire genomic alterations typical for fully metastatic cells later.58 Although surprisingly characterized by normal karyotypes, the malignant origin of a DTC population isolated from BM of breast cancer patients was demonstrated by other genetic alterations, such as allelic losses, and gene amplifications.59 While DTC can loose their ability to produce PSA, a significant percentage of DTC in prostate cancer patients still expressed PSA.60–65 LOH (loss of heterozygosity) pattern of PSA-positive CTC isolated from blood of prostate cancer patients indicated that CTC derive from distinct foci within the primary tumors.66 Moreover, K-ras mutations although not necessarily identical to those found in the corresponding primary tumors were reported in DTC derived from colon cancer.67, 68

DTC are heterogeneous with regard to the expression of growth factor receptors, adhesion molecules, proteases and their inducer and receptors, major histocompatibility complex antigens, signaling kinases, melanoma associated antigens (MAGE) or telomerase activity.1, 2, 63, 69–79 Of particular importance is the epidermal growth factor receptor HER2, the expression of which in primary tumors is essential for Trastuzumab treatment decisions of breast cancer patients.80, 81 HER2 overexpression on DTC in BM was predictive for a poor clinical outcome of stage I–III breast cancer patients.82 In the study of Vincent-Salomon et al.,83 both negative and positive HER2 status remained, in the majority of the cases, stable between DTC and the corresponding primary tumors. However, there is also evidence for discrepancies between the HER2 status in primary tumors and DTC in BM82, 84, 85 and CTC in blood.86, 87 Although HER2 expression was heterogeneous in DTC from individual patients, HER2-positive DTC might identify additional patients who can benefit from Trastuzumab therapy.85 Further studies are ongoing to verify whether HER2-positive DTC have predictive value for an improved response of the patients to Trastuzumab treatment.

Negativity for Ki-67 staining pointed to the hypothesis that the vast majority of DTC persists in a nonproliferating dormant state,84, 88, 89 which might also be responsible for their partial resistance to adjuvant chemotherapy in high risk breast cancer patients.14 In contrast using MIB immunostaining, Bianco et al. found in 36% of DTC-positive prostate cancer patients proliferating DTC the presence of which revealed to be an independent predictor of disease recurrence.90 Disturbance of dormancy and entry into the dynamic phase of metastasis formation might be caused by both genetic and epigenetic changes in DTC as well as in the surrounding microenvironment or premetastatic niche.91–93 However, conditions and timing of outgrowth of DTC are largely unknown thus far.1, 2, 11

There is first evidence for a molecular signature of primary tumors spreading early into BM94, 95; however, information about global gene expression of DTC is still limited. For CTC isolated from blood of patients with metastatic colorectal, prostate and breast cancer, global gene expression profiles could be defined and a list of CTC-specific genes was obtained, which might be useful to distinguish normal donors from cancer patients.96 Furthermore, Smirnov et al.96 concluded that gene expression profiles for CTC may be used to differentiate among the 3 different metastatic cancers. Interestingly, TWIST1, a transcription factor that plays a pivotal role in metastasis by promoting epithelial-mesenchymal transition,97–100 was part of the gene expression signature identified in EpCAM-enriched cells from BM of breast cancer patients after chemotherapy.101 Not observed in EpCAM-enriched cells of BM from healthy volunteers, TWIST1 expression was associated with occurrence of distant metastasis and local progression, even in pretreatment BM samples.101

Whether DTC possess stem or progenitor cell features enabling them to both self-renew and differentiate is currently under debate.102–105 Among the various breast cancer stem cell markers, CD44 positivity and absent/weak expression of CD24 or expression of ALDH-1 seem to be characteristic for breast cancer “founder” cells with a high capacity to form tumors in immunosuppressed mice.106–111 Recently, by gene expression analysis of CD44+/CD24−/low cells separated from breast cancer tissue, a 186 gene “invasiveness gene signature” was published, which was significantly associated with a reduced overall and metastasis-free survival of the breast cancer patients analyzed.112 Furthermore, different developmental pathways such as Notch, Wnt and hedgehog have been described to play pivotal roles in regulating cancer stem cell features.113–116 Thus, several sets of genes and proteins have been identified which might be helpful for the further characterization of DTC regarding their potential stem cell-like properties. Hints for stem cell features of DTC in BM were provided by Balic et al. and Alix-Panabieres et al. who demonstrated a significant number of DTC from BM of breast cancer patients with either CD44+/CD24−/low or CK19+/Muc-1 stem cell-like phenotypes.25, 117 The nonproliferating state of DTC that renders them resistant to systemic chemotherapy and the long term persistence of these cells in BM of cancer patients is also consistent with their putative stem cell phenotype.14, 16, 118

Clinical Relevance of DTC

Several studies have shown significant correlations between the presence of DTC in BM and metastatic relapse in various tumor types, suggesting that the founder cells of overt metastases might be among those DTC. Here we will briefly review the current state of knowledge, focusing on the main types of solid tumors (i.e., breast, colorectal, nonsmall lung, and prostate carcinomas). Additional studies have been performed in patients with other epithelial tumor entities, such as gastric cancer,119 esophageal cancer,120, 121 pancreatic cancer,122, 123 ovarian cancer and head and neck carcinomas.124–131

In colorectal cancer, approximately 50% of patients undergoing a curative resection (R0) die from metastatic disease within 5 years. Even among lymph node negative (N0) patients, the relapse rate is 30%.132, 133 In lung cancer, the prognosis is even worse, with 60% of R0 and 40% of N0 patients dying of the disease.134 Whereas in breast and prostate cancer the overall survival is today relatively high (80–90%, 5 year; 70–80%, 10 year), a considerable fraction of nodal negative patients still relapse (25–30% and 15–50%) and this can often take place many years (>10 years) after the removal of the primary tumor.135–137

Breast cancer

Numerous studies have been published investigating the presence and clinical relevance of DTC in BM of breast cancer patients.1, 15, 138, 139 The size of the patient cohorts analyzed and the applied detection methods vary considerably between these studies (Table I). The most commonly used method for DTC detection in breast cancer is immunocytochemistry-based analysis using antibodies against either cytokeratins or mucins (e.g., MUC-1). In most studies detecting these antigens, the prevalence of DTC in early-stage disease ranges between 13 and 50% (Table I), whereas in patients with metastatic breast cancer the rate increases up to 70%.138 RT-PCR-based methods have mostly used CK19, CEA or mammaglobin as markers. The detection rates between these studies varied between 12 and 50% (Table I) and raising up to 80% in metastatic breast cancer patients.138, 139

Depending on the method used and the number of patients analyzed, the different studies have come to somewhat different conclusions regarding the clinical relevance of DTC (Table I). However, Braun et al.15 published a meta-analysis consisting of 4,703 breast cancer patients. In this pooled analysis, the presence of DTC in BM was not only predictive of the development of skeletal metastases but was also predictive for the development of metastases in other organs.15

Besides their presence at primary diagnosis and surgery, DTC have been described to survive chemotherapy and hormonal therapy,14, 88, 140 and they can persist in BM over many years post-surgery. This persistence is also linked to an increased risk of late metastatic relapse.16, 54, 118, 141, 142 For example, in high-risk breast cancer patients (>3 involved axillary lymph nodes or extensive invasion of cutaneous lymph vessels), the presence of tumor cells after therapy was associated with an extremely poor prognosis.14 A European pooled analysis involving 696 breast cancer patients confirmed these findings. Here, 16% of breast cancer patients had tumor cell persistence in BM, and their presence was an independent prognostic factor for subsequent reduced breast cancer survival.143

Colorectal cancer

Most studies describing immunocytochemistry for the detection of DTC in colorectal cancer (CRC) patients have either used the monoclonal antibody CK2 against CK18 or the pan-cytokeratin antibody A45-B/B3 (Table II), and a few studies have used a cocktail of several (up to 5 different) antibodies against different epithelial antigens including cytokeratins.144–147 The DTC detection rate in studies using CK2 was between 16 and 32%, whereas the detection rate was clearly higher in studies using the A45-B/B3 antibody (24–55%). This difference might be due to a potential downregulation or loss of CK18 expression on DTC.18, 148 Both antibodies rarely detect CK-positive cells in BM of noncancer control patients (0–5.5%). In studies using a cocktail of different antibodies, the DTC detection rate in cancer patients has varied between 29 and 74%.

The largest study so far conducted has been performed by Flatmark et al.149 including 275 patients and 206 noncancer control patients. In this study, an immunomagnetic bead enrichment system capturing DTC by the antibody MOC31 directed against EpCAM was applied. More BM cells could be analyzed (2 × 107) than in other studies and 17% of the patients showed DTC, whereas only 1.5% of controls were positive. The BM status did not correlate significantly with disease stage or other clinical parameters. Unfortunately, no clinical follow-up was available, and the clinical significance of the findings in cancer patients remains therefore unclear.

So far 5 publications have reported a positive association between DTC in BM and an increased recurrence rate150, 151 or a reduced overall survival.145, 152 In contrast, 3 other reports, mostly on smaller patient cohorts, could not detect any association between prognostic factors and the presence of DTC.146, 147, 153

To date, CK20 is the most commonly used marker in RT-PCR analyses of BM from CRC patients (Table II), with more than 10 studies published so far. However, only in 3 studies the number of investigated patients was larger than 100 with DTC detection rates of 11–35%.154–156 The detection rate of CK20 in healthy controls ranged between 0 and 10% except for 1 study with 45% positivity.156 Only very few studies have investigated the correlation between CK20 transcripts in BM and prognosis. Two groups reported no association to survival,157, 158 whereas 4 groups found an association between the presence of CK20 transcripts and worse overall survival.154, 159–161 However, both negative reports were performed on metastatic patients only.

CEA (carcinoembryonic antigen) transcripts were also searched for in 1 small study involving only 10 patients with 66% of whom had detectable CEA transcripts, whereas none of the 56 healthy controls had CEA-positive cells in their BM.162 No follow-up information was available for these patients. In 1 other study, both CK20 and GCC (guanylylcyclase C) were investigated simultaneously.156 While CK20 transcripts could be detected in 11% of CRC cases, GCC mRNA was only detected in 6% of the patients. No GCC expression was found in healthy controls. Interestingly, only 1/109 patients showed simultaneously GCC and CK20 transcripts. None of these transcripts correlated with clinical parameters such as stage or grade and again no survival data were reported.

In conclusion, several studies have been conducted to elucidate the clinical relevance of DTC in CRC. These studies, however, present a very heterogeneous picture, with different patient groups, sample sizes, follow-up times, staining methods, and target antigens, all of which probably have contributed to the observed variation in DTC detection rates and association to clinical parameters (Table II).

Lung cancer

Table III shows the 13 ICC studies investigating the prognostic relevance of DTC in BM of patients with nonsmall cell lung carcinomas (NSCLC). In 7 studies, the monoclonal antibody CK2 against CK18 was used, whereas 5 studies have exploited different pan-cytokeratin antibodies. The rate of CK-positive cells in the different studies ranged between 22 and 60%. While the frequency of positive patients does not seem to correlate with the antibody used, the sampling number and also the anatomic site of BM aspirations might influence the number of positive cases. A higher frequency of DTC was not only found in BM aspirations from the rib or sternum in lung cancer patients163–165 but also among breast and oesophageal cancer patients when compared with BM samples taken from the iliac crest.166, 167

Irrespective of the localization of the BM puncture, 8 studies have shown a correlation between DTC in BM and worse clinical outcome.163–165, 168–172 However, the largest study conducted so far on 196 patients could not find an association, which can, however, be explained by a very short follow up time (median, 8 months).173

Interestingly, very few studies have been performed using RT-PCR to detect DTC in BM. Sienel et al. detected MAGE-A transcripts in 50 NSCLC cases free of overt distant metastases.79 Twenty-two percent of the patients harbored MAGE-A transcripts and such transcripts were undetectable in healthy controls. Among lymph node-negative patients, the presence of MAGE-A was associated with poor prognosis and revealed to be an independent risk factor in a multivariate analysis. In another smaller study, blood and BM of both SCLC (small cell lung cancer) and NSCLC cases were investigated for the presence of gastrin-releasing peptide (GRP). While no transcripts could be detected among NSCLC cases, in 50% of the blood and 18% of BM samples from SCLC patients GRP mRNA was detected.174

In conclusion, around half of the published studies on the prognostic role of DTC in BM in nonsmall cell lung cancer have found an association between the presence of DTC and worse overall survival or recurrence rate. However, most studies conducted so far have only investigated a small number of patients, and thus larger standardized studies are needed to confirm the prognostic value in lung cancer.

Prostate cancer

Currently, prostate cancer biology is only incompletely reflected by the commonly available diagnostic/prognostic parameters. Consequently, additional diagnostic parameters are urgently needed. BM is the most prominent metastatic site in prostate cancer and several research groups have focused on the detection of DTC in this organ over the past 10 years (Table IV). However, most of the studies included a relatively small number of patients and/or follow up information, and data about the influence of hormonal treatment on prognosis are sparse or lacking.

Immunocytochemistry using anti-cytokeratin and anti-PSA antibodies was the most frequently conducted approach (Table IV).175–179 DTC detection rates ranged between 10 and 90% (Table IV) and a comparison between the results obtained with the CK2 (anti-CK18) and the pan-CK antibody A45-B/B3 revealed a downregulation of CK18 of DTC in a significant number of cases.148, 180

There is evidence for a correlation of DTC to clinically established risk factors. Spread of DTC seems to be linked to histological differentiation of the primary tumor.61, 181 Moreover, while in some studies DTC detection was not associated with initial PSA concentration in the serum of prostate cancer patients,177, 179 a correlation of DTC measurement to early PSA recurrence was shown by 2 studies.179, 180 Weckermann et al.180 analyzed 82 patients and observed that the mean time to PSA increase was shorter in patients with DTC-positive compared with DTC-negative BM samples. As shown by Bianco et al.,90 the presence of MIB-1-positive proliferating DTC in bone marrow was associated with increased postoperative Gleason score.

Although a DTC-positive BM status was associated with grading and increased risk of metastasis, the study by Berg et al. (2007)181 on 266 patients did not find a correlation of DTC detection and survival. Most recently, Köllermann et al179 reported on the prognostic relevance of DTC in BM found in 86/193 (44.6%) patients with clinically localized prostate cancer submitted to neoadjuvant hormonal therapy followed by radical prostatectomy and a median follow-up of 44 months. This study is the first one on a larger series of prostate cancer patients with sufficiently long follow-up, which clearly demonstrates an independent adverse prognostic effect of the presence of DTC at the time of initial diagnosis. The presence of DTC in BM of prostate cancer patients found by A45-B/B3 ICC before neoadjuvant hormonal therapy and subsequent surgery represented even an independent prognostic parameter, suggesting that DTC may contribute to the failure of current neoadjuvant hormonal therapy regimens. Addition of information about DTC detection in BM to the conventional diagnostic parameters might lead to more precise reflection of the biological potential of an individual prostate carcinoma.179

There are also studies that used RT-PCR for DTC detection, mostly amplifying PSA- or MAGE-specific cDNAs as markers with detection rates ranging between 16 and 83% (Table IV). The BM status in these studies frequently correlated to PSA serum levels; however, there are only sparse data about the prognostic relevance or potential predictive values of these findings. Results from BM analysis presented by Wood and Banerjee176 demonstrated a correlation between RT-PCR-PSA positivity and a significantly shorter disease-free survival. Furthermore, RT-PCR status correlated significantly with serum PSA and pathological stage.176 In the study by Gao et al.,182 no association between BM RT-PCR PSA positivity and grading or serum markers was observed, but PSA mRNA detection correlated significantly with early disease recurrence.

In conclusion, there is some evidence that the detection of DTC in the BM of prostate cancer patients might represent a prognostic parameter. DTC appear to survive hormonal therapy183, 184 and this might be one explanation for the well-known failure of neoadjuvant hormonal therapy. However, improved characterization of DTC and larger multicenter studies followed by nomogram testing against the established risk parameters are required to introduce DTC detection into the future clinical management of prostate cancer patients.

Conclusion and future directions

The development of rare cell detection techniques allows now the reliable detection of DTC in BM of cancer patients years before the occurrence of distant overt metastases signals incurability. This information may be used to assess the individual prognosis of cancer patients and stratify the patients at risk to systemic adjuvant cancer therapies aimed to prevent metastatic relapse. However, for future implementation of DTC detection into clinical practice, it is essential to establish standardized methods with a high degree of reproducibility. An important step into this direction has been thus far made for the immunocytochemical detection of DTC using anticytokeratin antibodies.17, 20–22

Currently, most data on the prognostic value of DTC are available for breast cancer. In other tumor entities, the database is not as solid and the results are more controversial as in breast cancer, but several single institution studies including patients with colon, lung and prostate cancer have documented a link between the presence of DTC at primary surgery and subsequent metastatic relapse. Multicenter analyses are now required to confirm the prognostic value in these tumor entities.

DTC in BM have been detected in all solid tumor types suggesting that the BM might be a preferred reservoir for blood-borne DTC. The BM environment may allow these cells to persist over many years and to disseminate into other organs. Whether DTC use for this purpose the same niches as haematopoetic stem cells185 is subject of current investigations. However, it cannot be excluded that BM is simply a convenient indicator organ and that early DTC are also present in other organs such as lung or liver, which are less easily accessible than BM. Rapid autopsy programs have the potential to address this important question.

Several studies found DTC in BM even many years after surgery and adjuvant therapy, and the presence of these DTC was associated with an increased risk for metastatic relapse.16, 54, 118, 120, 141, 142 These results demonstrate that DTC can reside in a prolonged latent state referred to as dormancy before they grow out into overt skeleton metastases. To understand this peculiar stage of dormancy together with the identification of the founder cells of overt metastases (“metastatic stem cells”) are some of the most important and challenging areas of basic research on the biology of DTC. In this context, the development of appropriate mouse models, which mimic the stage of minimal residual cancer, are of utmost importance to perform functional studies for validation of the descriptive findings observed in cancer patients.

Sequential peripheral blood drawings in particular for real-time monitoring of minimal residual disease in cancer patients undergoing systemic therapies (e.g., chemotherapy, hormonal therapy or antibody therapy) should be more acceptable than repeated BM aspirations. To date, it is not clear if CTC measurements could replace the examination of BM. To answer this question, many research groups are currently assessing CTC in clinical studies with encouraging results (for review, see Ref. 9). Implementation of CTC/DTC measurements into clinical trials has the potential to obtain an important future biomarker for real-time monitoring of the efficacy of systemic adjuvant therapy in individual patients.