Micrometastatic, circulating tumor cells

Written by on

I would like to share with you information from Prof. Dr. Michael Giesing, MD, PhD, the molecular geneticist, physician who, in the late 1970′s, and 80′s put together a test which isolated circulating tumor cells from the blood of cancer patients, identified the cancer genetics of these cells then tested them pharmaco-genetically to determine which cancer therapies worked and which did not. He did his graduate work at Columbia University in NYC then worked for the U.S. National Institute of Health (NIH). The TIRNA or tumor induced RNA test was developed by Dr. Giesing through the NIH. When he introduced this test, Dr. Gieseng states that the NIH acknowledged the importance of such technology but did nothing to develop it. Thereafter, Dr. Giesing returned to Germany to academic life. In the late 1990′s and 2000′s, Dr. Giesing started a lab to introduce his oncogenetic test for circulating tumor cells to the public. The following notes are Dr. Giesing’s rational for using circulating tumor cell molecular analysis and why it is superior to genetic analysis of solid tumor. My thanks to Dr. Giesing for advancing this technology which is currently in clinical trials at Harvard Medical College, MD Anderson Medical Center and Sloan Kettering Cancer Center under the corporate sponsorship of Johnson & Johnson, Inc., over thirty years after it’s initial creation. So goes the nature of cancer medicine in the US.

Introduction: The role of disseminated, latent micrometastases in patient`s safety and disease manangement
Michael Giesing, M.D., Professor of Biochemistry, University of Bonn – Germany, Head of Onkokonsult Laboratory, Zum Herzfeld 5, D-49536 Lienen – Germany

The personal shock being told of a malignancy hits the patient in a situation of devastating helplessness. At this point the medical society envisages the obligation to offer help in the areas of surgery, systemic treatment and psychology. Whereas the usefullness of high caliber surgery and psychology is free of doubt systemic treatment still lacks efforts to identify the susceptible patient prior to treatment, to control drug efficacy prior to the metastases disaster and to detect early drug resistance factors. Fullfilling these needs requires other measures to be taken than watchfull watching with occasionally inserted imaging procedures. It should be taken into consideration that a single and unvisible cancer cell left in the body after primary intervention may cause tumor relapse. The entity of such cells is defined as “minimal residual cancer cells” (syn.: MRD cells, disseminated latent micrometastases and others). These cells can be detected in blood and bone marrow by molecular techniques.
Genetic profiling of MRD cells (DNA; RNA) with PCR based techniques has distinct advantages over the exclusive view on tumor masses, because these cells are the missing link between the primary tumor and possible metastases formation (see Fig.1) among them:

  • Early detection of the “at risk” patient
  • Independent “on line” prognostica-tion of the disease course
  • High prediction value for systemic treatment
  • Drug monitoring
  • Early detection of drug resistance
  • Adjunct treatment of resistance factors

Whereas tumor tissue analysis is a single diagnostic event requiring profiles in courage to prognosticate individually for the whole life span MRD cells are dynamic, i.e. treatment regimen leave a specific molecular signature unless the blood / bone marrow is cleared from them. MRD cells may also switch from the deposite, quiescent form in bone marrow to the more active form in the blood stream or vice versa. Worldwide acceptance of MRD cells as the prequisite for metastasis formation has led to a new view on cancer namely regarding the disease from a locoregional and a systemic perspective. Needless to say that MRD analysis ought to be introduced in cancer disease management on a broader scale.
The following short articles have been designed to give closer information on the practical use of MRD cells in patient management, choice of treatment (Pharmagogenomics), clinical research and drug development.

Why anti-cancer drugs work or fail: studies on the role of disseminated, latent micrometastases in disease monitoring and predicting therapy outcome

Michael Giesing, M.D., Professor of Biochemistry, University of Bonn – Germany, Head of Onkokonsult Laboratory, Zum Herzfeld 5, D-49536 Lienen – Germany

Risk assessment and therapy decisions in patients with solid tumors are routinely made after examination of the primary tumor, i.e. the TNM Status and –more recently- geno- and phenotyping. This method does not take into consideration minimal residual disease cells vagabonding in the patients` blood/bone marrow although these are actually therapeutic target cells. Obviously, anti cancer drugs should prevent latent micrometastases from progression into ouvert metastases. Also, predicting lifelong therapy outcome from a single analysis of the heterogeneous tumor tissue can be achieved only on statistical grounds whereas cellular disease monitoring in blood or bone marrow is an „online“ format designed to achieve this goal individually for each patient. In view of this challenge we and other groups have studied the role of blood borne, latent micrometastases for early and independent detection of metastatic risk, therapy susceptibility and resistance associated factors and for serving as a novel, additional therapy decision platform. We report here data from female breast cancer patients.

Disseminated, latent micrometastases: Applying our proprietary technology we can show solid tumors disseminate cancer cells in two forms namely M1 cells and M0 cells. The malignant nature has been previously shown by PCR-based typing of DNA aberrations in isolated cancer cells in comparison to MNC, i.e. diseased vs. non diseased. For routine purposes we have developed the PCR based TIRNA Test (Tumor cell Indicator RNA) derived from the capability of disseminated tumor cells to express high levels of oxidative stress genes (SOD, GPX1, TRNRD1; see Fig.3). Whereas M1 cells are almost devoid of epithelial expression characteristics (EGP, Cytokeratin) epithelial M0 cells harbour still organotypic messages suggesting two different microtumor entities. M1 cells are premetastatic cells, characterized by high dedifferentiation. M0 cells, however, may accumulate metastogenic potential through epithelial-mesenchymal transformation converting into M1 cells.. This has been shown by prospective observation of relapse-free disease. The phenotype of both cancer cells forms may be different from the primary tumor (Table I). This has a fundamental impact on therapies since disseminated cancer cells are the leading target for drug therapies..

Disease monitoring: Anti cancer therapies leave specific molecular signatures in M1 and M0 cells unless cells are eradicated by drugs. Hence, successfull therapies either clear the blood from M1 cells or reduce the number of DNA aberrations or induce a favourable phenotype in a dosage dependent manner. In contrast therapy failures induce expression of resistance associated factors thus increasing the metastatic risk (Fig.4). The Phenotype profile in M1 cells isolated from blood after successful therapy is shown in Fig. 5. The patient underwent inadequate surgery presenting high TIRNA and CK20 values in disseminated cancer cells. Drugs were tested. Topo I Blockade with Topotecan and nucleic acid inhibition with Xeloda were tested sensitivie (analytical panel not shown). Following treatment reanalysis was performend. The blood was almost cleared from M1 cells; Drug targets and down stream effectors were inhibited; drug resistance factors could not be detected (not all data shown); clinically the patient showed complete remission of the tumor mass. The M1 test system indicates a strongly reduced risk for disease progression provided the drug effect is lasting.

Molecular drug sensitivity testing and therapy outcome prediction: The possibility associated with the M0 / M1 System of detecting patients “at risk” in an “on line” fashion and to monitor drug activity in target cells prompted us to develop molecular molecular assays for drug sensitivity testing and predicting therapy outcome. Our efforts culminated in test systems measuring in a drug specific manner the pathways a drug must affect to induce apoptosis in target cells (Fig. 6). Factors exerting affects in patient`s survival play a pivotal role in this concept. Thus we could realize the pharmacogenomics principle “right drug to the right patient in the right dosage at the right time”. Following this approach we were able to circumvent the vagaries associated with drug incubation of target cells. The principles of Pharmacogenomics additionally habours the therapeutic option to suppress resistance factors making drug regimen more successful.

In various clinical studies the technique of molecular typing of M1 and M0 cells for predicting therapy outcome at the metastatic stage was used. Regardless of the molecular drug sensitivity classification the oncologists were free to chose their drug regimen. The clinical outcome was tested by imaging procedures and –wherever possible- by molecular reanalysis of M1 / M0 cells. Also molecular data have been compared with clinical imaging results (Fig. 7). Drugs being tested sensitive led to stable disease or tumor remission in 84% of patients. Application of drugs tested resistant led to progressive disease in 77%. The data render to M0/M1 cells a predictive factor of 81% as calculated from the clinical data. Due to the extreme heterogeneity comparative analysis in ouvert clinical tumors a much lower predictive value was found. It is noteworthy to acknowledge that the relaspe-free survival as well as the overall survival of solid tumor patients depends on M1 cells rather than on the size of clinical tumors. Clearing the blood stream from M1 cells was found to code for a better disease outcome.

Conclusions: Application of a powerfull geno-/phenotyping in latent micrometastases is of considerable relevance for both oncologists/patients and drug suppliers since the disease is classified rather from malignancy (M1/M0) than from location. Patients being treated after the analytical set of predictive genes detected in disseminated micrometastases live longer without disease progression. Reading out the appropiate drug dosage is done routinely. Our data clearly show that most cancer patients need more than one single treatment regimen. M1/M0 typing is a rational for sequential therapy selection. Drug development can be achieved much faster than before through studying drug targets and pathways in these cells in comparison to tumor tisssues. In clinical studies new patient stratification as well as early therapy monitoring may become routine. We suggest to industry to make use of archived micrometastases for early differentiation between candidate responder patients and non-responder patients.

Selected Literature:

1) Austrup F, Uciechowski P, Eder C, Böckmann B, Driesel G, Jäckel S, Kusiak I, Grill H-J, Giesing M. Prognostic value of genomic alterations in minimal residual cancer cells purified from blood of breast cancer patients. BJC 2000; 83(12): 1664-1673

2) Behrendt W, Uciechowski P, Kusiak I, Giesing M. Die Behandlung von disseminierten Mikrometastasen bei Patientinnen mit Mammakarzinom durch unspezifische Immunmodulation. Curr.Oncol. 2001; 11(2): 31-42

3) Braun S, Marth c: Circulating tumor cells in metastatic breast cancer – towards individualized treatment? N Engl J Med 2004; 351: 824 – 826

4) Cristofanilli M, Budd GT, Ellis MJ, Stopek A, Matera J, Miller MC, Reuben JM, Doyle GV, Allard WJ, Terstappen LWMM, Hayes DF. Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 2004; 351: 781-791

5) Berber B, von Minckwitz G, Raab G, Schütte M, Hilfrich J, Blohmer J-U, Costa S, Eidtmann H, Jakisch C, du Bois A, Kaufmann M. Stellenwert der primären Chemotherapie beim Mammakarzinom. Dtsch Ärztebl 2003; 100: A 2378 – 2382 [Heft 37]

6) Giesing M, Austrup F, Böckmann B, Driesel G, Eder C, Kusiak I, Suchy B, Uciechowski P, Grill H-J. Independent prognostication and therapy monitoring of breast cancer patients by DNA/RNA typing of minimal residual cancer cells (MRCC). Int J Biol Markers 2000; 15(1): 94-99.

7) Giesing M, Grill HJ, Haselhorst U. Früherkennung solider Tumoren: Der Minimale Tumor. Curr.Oncol. 2001; 11(1): 31-38

8) Giesing M, Prix L, Uciechowski P, Suchy B, Eder E, Böckmann B, Driesel G, Kusiak I, Schütz A. Frühdiagnose und Therapiesensitivitätstests solider Tumoren mit Hilfe gentechnischer Verfahren. Curr.Oncol. 2001; 12(1): 32-44

9) Goetz MP, Ames MM, Weinshilboum RM. Primer on medical genomics. PART XII: Pharmacogenomics – general principles with cancer as a model. Mayo Clin Prog 2004; 79(3): 376-384

10) Robert J, Bonnet J. Recent advances in pharmacogenomics in oncology. Bull Cancer (France); 2004; 91(1): 19-28

11) Slonim DK. Transcriptional profiling in cancer: the path to clinical pharmacogenomics. Pharmacogenomics (England); 2001; 2(2): 123-136

12) Schmidt-Kittler O, Ragg T, Daskalakis A, Granzow M, Ahr A, Blankenstein TJF, Kaufmann M, Diebold J, Arnholdt H, Müller P, Bischoff J, Harich D, Schlimok G, Riethmüller G, Eils R, Klein CA. From latent disseminated cells to overt metastasis: Genetic analysis of systemic breast cancer progression. PNAS 2003; 100 (13): 7737 – 7742

13) Schumacher K, Stoll G. Das Integrative Konzept Onkologie. Deutsche Zeitschrift für Onkologie 2003; 35: 37-51

14) Weinstein JN. Pharmacogenomics- teaching old drugs new tricks. N Engl J Med; 2000; 343(19): 1408-9

What Does Cancer Mean?

What does cancer mean: only a solid tumor mass or a systemic disease?

Michael Giesing, M.D., Professor of Biochemistry, University of Bonn – Germany, Head of Onkokonsult Laboratory, Zum Herzfeld 5, D-49536 Lienen – Germany

Hematologic malignancies have been known for a long time to disseminate cancer cells into the blood stream. Theses cells are used for disease monitoring in blood and in bone marrow. Disseminated cancer cells also serve diagnostic purposes in lymphomas. Identification of such cells is achieved through molecular analysis, e.g. gene translocation, gene rearrangement or others.

Solid tumors are by far the largest portion of malignancies. The rate of distant metastases formation is quite considerable in spite or all progress obtained by an increasing number of drugs. The TNM classification is the link among oncologists worldwide for understanding the disease. This view, however, does not take into consideration the dynamics of the disease because it´s merits is restricted to a single point analysis making individual prognosis over a life span impossible. Decades of experience with the TNM system have generated a body of statistical data which –in spite of its volume- can hardly predict individually therapy outcome. Statistics, however, means for each patient to address individually the winner-looser group question. Answering this fundamental question becomes even more complicated looking on the percentage figures given by clinicians for risk or therapy success. In contrast to mathmatical rules clinical percentages are routinely expressed relatively to a favorable subgroup but not to the total disease population as necessary. Thus the success figures may be misleading and hinder clinicians to achieve better results through targeted sequential therapies.

The oncologist´s view on tumor mass has entirely neglected the question what mechanism and what kind of stuff may be responsible for disease progression even after complete surgical removal of the tumor and systemic medication. Among the established surrogate markers most tumor serum proteins have failed to act as that stuff. Today there are quite some hypotheses what kind of stuff stands for disease progression, e.g. soluble tumor nucleic acids infecting distant organs and causing a transition into malignancy. Another hypotheses underlines disseminated cancer cells causing metastases formation.

Oncoconsult´s proprietary technology envisages the role of disseminated micrometastases in some of the urgent questions related to cancer patients` management (Fig.2). Onkokonsult has collected great experience in analysing disseminated micrometastases isolated from blood and bone marrow in the last 12 years. Reviewing the considerable number of results and many years of prospective observation Onkokonsult has created the DISCellâ -Paradigm based on PCR-based geno- and phenotyping of disseminated cancer cells. The results obtained from thousands of patients followed by clinical long term observation has tought us that cancer is not only a locoregional episode. In contrast the data allow us to make the clear statement “Cancer is a systemic disease” evolving from the following key findings which stand for a novel approach in patient management:

  • Patients with solid tumors disseminate various cancer cell types (DISCCellÒ: M0 and M1Type)

    • M0 Type micrometastases still show up with organotypic gene expression and have little metastogenic potential; there are exceptions from this rule

  • M1 Type is a dedifferentiated metastastogenic cancer cell
  • The prevalence of tumor cell dissemination varies among tumor entities reaching from <50% to >90%.
  • The geno- / phenotype of disseminated micrometastases is unlike the primary tumor
  • Micrometastases in bone marrow are often quiescent whereas the active cells circulate in the bloodstream
  • Relapse-free survival and overall survival is governed by the geno- and phenotype of DISCCellÒ
  • Disseminated micrometastases represent a novel tumor entity serving as the vital targets in systemic therapies
  • Different therapy classes leave specific molecular signatures in micrometastases
  • The majority of untargeted therapies fails to clear the blood from micrometastases
  • Over 60% of untargeted drug therapies leave micrometastases with drug resistence factors most of them contributing to an unfavorable prognosis
  • Resistance factors can be down regulated by adjunct therapies
  • Therapies may fail if micrometastases show up with drug targets and down stream effectors unlike the clinical tumor mass
  • Molecular drug sensitivity testing has a prediction value of 81% thus exceeding by far what can be reached in clinical tumors
  • The majority of cancer patients needs more than a single systemic treatment to avoid metastases formation

The role of disseminated, latent micrometastases in cancer drug development

Michael Giesing, M.D., Professor of Biochemistry, University of Bonn – Germany, Head of Onkokonsult Laboratory, Zum Herzfeld 5, D-49536 Lienen – Germany

Tumor drug development has become a costly enterprise requiring increasing amounts of financial investment albeit with uncertain success in the market. The former enthusiasm for targets and the enormous realm of HTP techniques has generated a mass of data which has the potency to neglect questions related to early target patient identification and market penetration. Early focussing on the real target cells and candidate patients is possible through the use of Onkokonsult`s DISCellâ proprietary system (Fig. 11). This analytical system involves geno- / phenotyping of disseminated latent micrometastases acting as a novel tool to identify safely drug susceptible patients.

Today the chance for bluckbuster drugs is limited and more or less replaced by niche products applicable only to a fraction of diseased patients. Also the mode of clinical research requiring an increase in life span of patients at a highly advanced cancer stage in order to obtain approval for a drug is not really helpful for daily drug application. Years of continuing clinical research pass until first line treatment is approved. Finally drug approval in a given tumor entity does not meet the idea of pharmacogenomics demonstrating individual drug efficacy in susceptibel target cells regardless of the primary tumor. Examples for this approach are Herceptin and Gleevec acting in more tumor entities than originally approved for.

The pivotal question for susceptible patients, however, will be adressed at the clinical level when years of drug development have passed and daily practice depicts a pattern of drug applicability unlike or in addition to the primary approval. The need for target patient identification has been already addressed by the FDA. In the future this clue to drug application will come even more into the focus of regulatory authorities as well as of reimbursements companies.

All attempts to lubricate the path of drug development and clinical profiling in solid tumors are hindered by the sole view on clinical tumor masses. Apart from watchfull waiting tumor sizing is the predominat on line approach to assess clinically the disease course. This is surprising since the cellular heterogeneity of clinical tumors has been known for a long time, so is the dissimilarity bewteen primary tumors and metastases. Also tumor size is hardly associated with overall survival. More recently disseminated cancer cells in blood or bone marrow came into discussion. Minmal residual disease cells –other synonyma are latent micrometastases- may play a pivotal role in the future since they are the intermediate tumor entity of solid tumors on their way to local disease expansion and distant neoformation of tumors. In fact these cells are the true drug targets in neoadjuvant, adjuvant and even palliative treatment settings and represent, therefore, the missing link between clinical tumors and course of the disease.

Onkokonsult´s proprietary technology envisages the role of disseminated micrometastases in some of the urgent questions related to faster drug development and to cancer patients` management. Onkokonsult has collected great experience in analysing disseminated micrometastases isolated from blood and bone marrow in the last 12 years. Reviewing the considerable number of results and many years of prospective observation Onkokonsult has created the DISCCellÒ -Paradigm (Fig. 12) based on PCR-based geno- and phenotyping of disseminated cancer cells. The following statements may be of relevance for drug development:

Patients with solid tumor disseminate various cancer cell types (DISCCellÒ: M0 and M1Type)
Relapse-free survival and overall survival is governed by the geno- and phenotype of DISCCellÒ
The majority of patients shows up with drug targets and the oncogenic phenotype of drug targets in M1/M0 cells respectively which is unlike the primary tumor
Prognosis of the disease course is independent from the ouvert tumor
Drug sensitivity testing has a prediction value for therapy outcome of 81% thus exceeding by far what can be reached in clinical tumors
Treatment monitoring as well as early drug resistance detection is routinely achievable in the DISCCellÒ- sytem
The majority of cancer patients needs more than a single systemic treatment to avoid metastases formation

Application of the DISCCellÒ sytem is possible at all stages of drug development. The key components to accelerate drug development and reducing costs are:

Building up CANAP (Central archive) from DISCCellÒ and / or tissue specimen for comparative analysis
Archiving specimen until analyses are needed
Determination of drug targets and drug pathways resulting in Kaplan-Maier diagramms
Identification of drug targets valuable to proceed in drug development at all stages
Identification of drug targets in DISCCellÒ from various tumor entities of interest
Application of the DISCCellÒ sytem in clinical research thus lowering the number of patients required for approval plus saving time

The role of disseminated, latent micrometastases in CAM therapies

Michael Giesing, M.D., Professor of Biochemistry, University of Bonn – Germany, Head of Onkokonsult Laboratory, Zum Herzfeld 5, D-49536 Lienen – Germany

Drug regimen in the area of complementary alternative medicine (CAM) have developed aside the usual clinical strategies. CAM therapies comprise a wide variety of different approaches including activators of the immunesystem, antioxidants, physical therapies and others. As far as stimulation of the immune system is concerned CAM therapies are believed to act in a nonspecific fashion which means that they may act primarily on natural killer cells (Fig. 8) rather than on cytotoxic CD8+ T-Lymphocytes. As shown in Fig. 9 CD8+ cells require highly differentiated cancer cells to confer the lethal message. Disseminated M1 cells, however, are undifferentiated. Also none of the CAM drugs has ever been shown to induce monoclonality in CD8+ cells necessary for the specific cytotoxic activity. This mechanistic consideration has a deep impact on how and when to treat cancer patients with CAM drugs. A wide realm of stuffs claims to strengthen the immunsystem among them all kinds of plants, plant extracts, algae and their extracts, fungi, bacteriae, viruses, vitamins, Carbohydrates, proteoglycans, amino acids, fatty acids and many other chemical entities. As in conventional cancer medicine the claims in cancer prevention and metastases inhibition associated with CAM therapies exceed by far the state of reliable data. However, the contrary conclusion to reject CAM therapies is as unjustified as awarding the philosopher`s stone of wisdom to conventional oncology.

Today the clue in cancer therapy is named “pharmacogenomics” meaning “the right drug to the right patient at the right time in the right dosage”. The future in oncology is identifying the susceptible patient and monitoring drug efficacy prior to tumor neoformtion.

The efficacy of CAM Therapies depends on the interaction bewteen NK cells and disseminated latent micrometastases of the M1 Type. On a laboratory scale (Fig. 10) we firstly isolate in our routine setting NK cells in a specific fashion and measure quantitatively the basal message levels of cytokines which may kill M1 cells. The stimulatory potency of NK cytokines is measured after IL-2 incubation, not after drug incubation. The drug action in vivo cannot be mimicked ex vivo. Secondly we isolate TIRNA positive M1 cells and quantify cytokine receptors and a transcription factor. This setting has led to the identification of the in vivo mechanisms of a variety of CAM drugs. We assume that other drugs claim similar mechanisms. The tests we have developed for CAM therapies are prognostic for the disease outcome. Also the tests have a predictive value of > 70% for CAM therapies.

It should be noted that CAM therapy must be given as long as time is required to achieve a lasting effect on clearing the blood from latent M1 cells. Remember that memory cells are not involved in CAM drug action. Needless to say that we have obeserved the appearance of drug resistance either on the NK side or on the M1 side or on both sides. This would require a change in drug regimen. Also we have identified M1 cells killing NK cells (“striking back”). We have tests for these circumstances as well. So far we have collected promising results for a targeted apllication of CAM therapy regimen. Up to 60% of the patients reach a partial temporary remission of M1 micrometastases. Among these we could identify a group of > 30% of patients in which targeted CAM therapy cleared the blood entirely from M1 micrometases resulting in metastases inhibition. Stable M1 disease can be observed in almost 20%. Another 20% of the cases CAM therapy activates M1 cells thus promoting metastases formation due to untargeted therapy or overdosaging. Problematic cases can be early identified through the test panel. In summary then CAM therapy if targeted is a usefull contribution to postpone or even lower the metastastic rate in cancer patients. Patients without clinical tumors but disseminating metastogenic latent and susceptible M1 micrometastases are candidate patients for a successful therapy.

Selected Literature:

  1. Behrendt W, Uciechowski P, Kusiak I, Giesing M. Die Behandlung von disseminierten Mikrometastasen bei Patientinnen mit Mammakarzinom durch unspezifische Immunmodulation. Curr.Oncol. 2001; 11(2): 31-42
  2. Bjelakovic G, Nikolova D, Simonetti RG et al. Antioxidant supplements for prevention of gastrointestinal cancers: asystematic review and meta-analysis. Lancet 2004; 364: 1219-1228
  3. Giesing M, Prix L, Uciechowski P, Suchy B, Eder E, Böckmann B, Driesel G, Kusiak I, Schütz A. Frühdiagnose und Therapiesensitivitätstests solider Tumoren mit Hilfe gentechnischer Verfahren. Curr.Oncol. 2001; 12(1): 32-44
  4. Schumacher K, Stoll G. Das Integrative Konzept Onkologie. Deutsche Zeitschrift für Onkologie 2003; 35: 37-51