Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, National Cancer Institute, NIH, Executive Plaza North, Rockville, MD, USA

Correspondence to: E A Sausville, Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, National Cancer Institute, NIH Executive Plaza North, Room 8018, 6130 Executive Blvd, Rockville, MD 20852, USA; Fax: 301 402-0831

This article is a 'United States Government Work' paper as defined by the US Copyright Act

As the drug discovery and developmental arm of the National Cancer Institute (NCI), the Developmental Therapeutics Program (DTP) plans, conducts and facilitates development of therapeutic agents for cancer and AIDS. DTP's goal is to turn 'molecules into medicine for the public health'. Areas of support by DTP are discovery, development and pathways to development for the intramural and the extramural community. The Developmental Therapeutics Program (DTP) operates a repository of synthetic and pure natural products, which are evaluated as potential anticancer agents. The repository derives from a historical database of greater than 600 000 compounds, which have been supplied to DTP from a variety of sources worldwide. The in vitro anti-cancer drug cell line screen established at DTP is unique in several respects. It has changed the NCI emphasis from a compound-oriented drug discovery effort to a disease-panel oriented exercise, emphasized human tumor cells derived from solid tumors, developed a high volume screening method that can adapt to processing of numerous chemical agents or natural source-derived extracts, that has minimized the use of animals, and saved on the amount of material required for the initial screening. The hollow fiber assay created at the DTP has demonstrated the ability to provide quantitative initial indices of in vivo drug efficacy, with minimum expenditures of time and materials and is currently being utilized as the initial in vivo experience for agents found to have reproducible activity in the in vitroanticancer drug screen. Drugs showing activity with unique mechanisms of actions are being further developed for treatment of hematopoietic neoplasms, prominent examples being flavopiridol, UCN-01 and depsipeptide among others.

Leukemia (2002) 16, 520-526. DOI: 10.1038/sj/leu/2402464

drug development; Developmental Therapeutics Program; drug discovery; molecular target

Overview of NCI drug discovery process

The discovery and development of novel therapeutic agents for the treatment of malignancies is of vital importance to cancer patients. The discovery of effective anticancer agents reflects a partnership of academic centers, industry and government. The National Cancer Institute (NCI) has played a pivotal role in cancer drug discovery and development. The philosophical approach to cancer drug discovery and development has evolved over several decades. Empirical approaches seek to define compounds or natural products (extracts of living organisms) which alter or inhibit cancer cell growth. Rational approaches start with a defined target structure or process and seek to design approaches to alter the target and hence tumor cell growth. In truth, all successful empirical approaches have ultimately had the goal of being understood in targeted or molecular terms. All 'rational' approaches have built on empirical observations of success in, for example, drug metabolism and distribution. Therefore a synthesis of aspects of the two approaches is ultimately desirable. The National Cancer Institute has a screening program that allows empirical evidence of antitumor activity to be rapidly understood in molecular terms that facilitate rational use of the screening studies.

The NCI's effort in new drug discovery and development grew out of a congressionally mandated initiative, the Cancer Chemotherapy National Service Center, which in 1955 established a national resource to facilitate the evaluation of novel chemicals as potential cancer chemotherapeutic agents. By 1976, its functions were incorporated into the Developmental Therapeutics Program (DTP) of NCI. DTP has had a role in discovering or further developing approximately 40% of the current USA-licensed chemotherapeutic agents.¹ The primary focus of NCI's pre-clinical drug evaluation program initially and to the present day is in the treatment of malignancy. However, the emergence of human immunodeficiency virus (HIV) also encouraged the involvement of NCI in the discovery and development of effective therapies for acquired immunodeficiency syndrome (AIDS).² Since the preclinical functions are essentially identical for both programs, resources already in existence for cancer drug discovery were mobilized for the development of therapies for HIV between 1988 and 1996 and two currently licensed HIV therapeutics emerged from that activity.

The present DTP organization (Figure 1) is made of 10 branches at locations in Frederick and Bethesda, Maryland. The DTP World Wide Web site (http://dtp.nci.nih.gov/) also gives access to the spectrum of its multitude of roles. The functions of the Developmental Therapeutics Program can be broadly divided into acquisition and discovery services, development and pathways to development, and information for the intramural and the extramural research community including pharmaceutical companies and academic institutions.

The Developmental Therapeutics Program (DTP) operates a repository of synthetic and pure natural products, which are evaluated as potential anticancer agents. The repository derives from a historical database of greater than 600 000 compounds, which have been supplied to DTP from a variety of sources worldwide. These materials, not based on 'proprietary' frameworks, represent unique structural diversity, as assessed by the Hodes model³ assessment of compound substructures. The collection contains both synthetic and fully characterized pure natural products. DTP's 'open' Synthetic repository contains more than 140 000 samples that have been submitted to the program on a nonproprietary basis for anticancer evaluation. These compounds are available to the extramural research community, details on the DTP World Wide Website (http://dtp.nci.nih.gov/), for the discovery and development of new agents for the treatment of cancer, AIDS or opportunistic infections afflicting patients with cancer or AIDS. Also available is the diversity set (1990 samples) available on request to peer-reviewed investigators 'no strings attached', which represent a systemic sampling of the structural diversity of the available repository, a mechanistic set (879 samples) chosen to represent a broad range of growth inhibition patterns in the 60 cell line screen, based on the intrinsic activity of the compounds, and the challenge set (57 samples) of novel structural types evaluated in the DTP human tumor cell line assay and have unusual patterns of cell line sensitivity and resistance, but currently unknown mechanism of action.

Since 1986, DTP has been acquiring plants and marine organisms through collection contracts performed in over 25 tropical and subtropical countries worldwide. As of September 1999, over 50 000 plant samples have been collected in Africa and Madagascar, Central and South America, and Southeast Asia through contracts with the Missouri Botanical Garden, the New York Botanical Garden, and the University of Illinois at Chicago, assisted by the Arnold Arboretum of Harvard University and the Bishop Museum in Honolulu. During the same period over 10 000 marine invertebrates and marine algae have been collected, mainly from the Indo-Pacific region, through contracts with the Harbor Branch Oceanographic Institute, the Australian Institute of Marine Science, the University of Canterbury, New Zealand, and the Coral Reef Research Foundation. Marine and plant species collection is still continuing through several of the above-mentioned contacts. NCI has committed itself to the conservation of biological diversity, as well as to policies of fair and equitable collaboration and compensation in interacting with the source countries participating in the collection programs. Agreements based on the NCI Letter of Collection have been signed with relevant government organizations in many of the source countries participating in the collection program. Each organism is extracted in the Natural products Extraction Laboratory with a 1:1 mixture of dichloromethane and methanol, and then with water, and the extracts are stored at -20°C, until tested in the NCI cancer cell line screen.

DTP maintains a preclinical repository of biological agents, including reference human and mouse cytokines for use as standards and calibration. Also available are monoclonal antibodies against human and mouse antigens for use in in vitro and in vivo basic research. NCI maintains a low-temperature repository of transplantable in vivo-derived tumors and in vitro-established tumor cell lines from various species, including human, guinea pig, hamster, mice, rabbit and rat. DTP's Angiogenesis Resource Center distributes endothelial cells along with reagents, standardized methods, and instructions on maintaining cells' performance in reproducible routine angiogenesis bioassays. Also available are angiogenic regulatory molecules, including proteins and small molecules that activate or inhibit endothelial cell migration and morphogenesis into new vessels.

Until 1985, the NCI screening program and the selection of compounds for further preclinical and clinical development under NCI auspices had relied predominantly on the in vivo L1210 and P388 murine leukemias and other transplantable tumor models.⁴ From 1975 to 1985, the in vivo P388 mouse leukemia model was used almost exclusively as the initial or primary screen. With few exceptions, agents that showed minimal or no activity in the P388 system were not selected by the NCI for further evaluation in other tumor models or alternative screens. Most of the available clinical anticancer agents are active in the P388 system. However, most were discovered prior to 1975 or by observation initially in test systems other than the NCI-operated P388 primary screen.

From 1985 to 1990 a shift in screening strategy was implemented, putting in place the concept of a disease-oriented in vitro primary anticancer drug screen as a replacement for the P388 in vivo primary screen.⁵ The new screen was unique in several respects; it changed the approach from a compound-oriented drug discovery to a disease-panel oriented exercise, emphasized human tumor cells derived from solid tumors, developed a high volume screening device that could adapt to either numerous chemical agents or extracts, and established an in vitro screen that would avoid the use of animals, and save on the amount of material required for the initial screening. The in vitro cell line screen was implemented in fully operational form in April of 1990 and can screen up to 20 000 compounds per year for potential anticancer activity. To date, more than 75 000 compounds have been tested. The operation of this screen utilizes 60 different human tumor cell lines, representing leukemia, melanoma and cancers of the lung, colon, brain, ovary, breast, prostate and kidney. Pertinent to the topic of this review, only six of the 60 are hematologic neoplasm, but they still allow valuable lead structure with hematologic neoplasm selectivity to be defined.³ The aim is to prioritize for further evaluation, synthetic compounds or natural product samples showing selective growth inhibition or cell killing of particular tumor cell lines. This screen is unique in that the complexity of a 60 cell line dose-response produced by a given compound results in a biological response pattern which can be utilized in computerized pattern recognition algorithms. Using these algorithms, it is possible to assign a putative mechanism of action to a test compound, or to determine that the response pattern is unique and not similar to that of any of the standard prototype compounds included in the NCI database. In addition, following characterization of various cellular molecular targets in the 60 cell lines, it has been possible to select compounds most likely to interact with a specific molecular target.⁶ Details of the screen can be examined on the DTP World Wide Website (http://dtp.nci.nih.gov/).

In early 1995, during the course of reviewing data from the cancer screen, it became obvious that many agents were completely inactive under the conditions of the assay. A protocol for a three cell line prescreen was developed, which would test for the presence of toxicity at 10^-4 M drug concentration and could eliminate a large proportion of the inactive agents, but preserve active agents for multi-dose 60 cell line testing. Computer modeling indicated that approximately 50% of drugs could be eliminated by this prescreen without a significant decrease in ability to identify active agents, and should be able to increase the output and efficiency of the main cancer screen with limited loss of information. The selected cell lines are MCF7 (breast), NCI-H460 (lung) and SF-268 (glioma). They are valid predictors for antileukemic activity because essentially 95% or greater of drugs active in these cell lines will have some degree of activity in hematopoietic neoplasm lines. Figure 2 illustrates the schema for screening compounds at the NCI drug discovery program.

Advancement of potential anticancer agents from 'discovery' in an in vitro screen to pre-clinical development requires a demonstration of in vivo efficacy in one or more animal models of neoplastic disease. Most such models require considerable materials in terms of laboratory animals and test compound as well as substantial amounts of time and cost to determine whether a given experimental agent or series of agents have even minimal antitumor activity. The hollow fiber assay⁷ created at the DTP has demonstrated the ability to provide quantitative indices of drug efficacy with minimum expenditures of time and materials and is currently being utilized as the initial in vivo experience for agents found to have reproducible activity in the in vitro anticancer drug screen. The correlation of hollow fiber activity with in vitro and in vivo activity has recently been published⁸ and utilized by corporate colleagues to assay effects on molecular targets.⁹ Standard panels of 12 tumor cell lines are used for the routine hollow fiber screening of the in vitro actives. In addition, alternate lines can be used for specialized testing of compounds in cell lines with a basis for selective action by the test drug. In a typical assay, cell lines in each of the two locations (subcutaneous or intra-peritoneal) are exposed to two dose levels of the test drug. A score of 0 is assigned for no activity, 1 for 1-50% reduction in MTT stain, and 2 for >50% reduction in all viability stain. The maximum possible score for an agent is 96. A compound is referred for xenograft testing if it has a combined i.p. (intra-peritoneal) and s.c. (subcutaneous) hollow fiber score of 20 or greater, a s.c. score of 8 or greater, or produces cell kill of any cell line at either dose level evaluated. This criterion would detect essentially all of the usually utilized 'standard' cytotoxic agents.

Numerous patterns of cellular responsiveness are generated by the many substances that are tested in the NCI cancer screen. Application of computerized programs and neural network analysis of the primary screening data have been utilized to refine interpretation of the screening results. For example, the computer program called 'COMPARE' uses a simple algorithm for comparing the patterns of cellular responsiveness for each compound against the extensive database that has resulted from the large number of past agents tested in the screen.¹⁰ This program is being utilized to identify possible mechanism of action of a new agent, survey those agents with similar mechanisms of action and also identify those agents that may truly have a unique mechanism of action. The computerized approach to analyzing the biological data might also be used to search for agents interacting with specific molecular targets. Although we continue to refer to this system as a 'screen', it also serves as a way to profile, or fingerprint, candidate therapeutic agents. Investigators at the NCI and other institutions are characterizing the 60 cell lines in terms of a variety of molecular factors at the DNA, mRNA protein, and functional levels. Among the targets assessed to date by various research groups are oncogenes, tumor suppressor genes, drug resistance mediating transporters, heat shock proteins, telomerase, cytokine receptors, molecules of the cell cycle and apoptotic pathways, DNA repair enzymes, components of cellular cytoarchitecture, intracellular signaling molecules, and enzymes of metabolism. Studying the multidrug-resistant (MDR) drug transporter by drug-response patterns in an attempt to understand the association between drug resistance and the MDR phenotype, exemplified the fact that compounds affecting common 'targets' would display specific patterns in the database related to the actual expression of the target. The similarity in pattern led to the identification and subsequent biochemical confirmation of many previously unrecognized compounds as substrates for the P-glycoprotein (P-gp) MDR drug transporter, indicating broader substrate specificity for P-gp than previously recognized.¹¹ The DTP World Wide Website (http://dtp.nci.nih.gov/) has information regarding MRPs and other transporters. Another target of great interest is p53 suppressor gene as most commonly used DNA-directed cytotoxic agents appear to be influenced by p53. In contrast, microtubule directed agents such as taxol, seems to be less dependent on p53 function.¹² This, detection of p53 'indifferent' compounds would be of interest. This approach has been extended to the correlation of RNA micro array expression data¹³ with drug action. Their results are available to the public through http://dtp.nci.nih.gov/.

An example of how this information can be correlated with clinical data emerge from a study where the 60 cell lines of the panel were assessed for the presence of a mutated ras oncogene, which was found to cluster in the leukemias, non-small cell lung and colon carcinoma sub-panels.¹⁴ A striking correlation was observed with cytosine arabinoside (Ara-C) sensitivity in cell lines harboring ras mutations compared with tumor lines with wild-type ras alleles. Similar correlations were observed with certain TOPO II inhibitors. The results suggest that the ras oncogene may play an important role in rendering tumor cells sensitive to certain agents. This hypothesis is supported by a clinical study of patients in response to treatment primarily comprising Ara-C and a TOPO II inhibitor.¹⁵

NCI and the Food and Drug Administration (FDA) in 1986 started an effort to evaluate the relationship between the preclinical toxicology and pharmacology of new anticancer agents. There has been a concerted effort to use the preclinical pharmacological data to expedite dose-escalation procedures during phase I clinical investigation of new anticancer agents. Preclinical toxicological studies allow definition of a safe initial dose for humans and a projection of the qualitative and quantitative toxic events that are likely to be encountered during the clinical trial. Efforts are directed to determine the doses of new drugs that cause lethality in 10%, 50% and 90% of the animals receiving the agent on varied schedules (ie LD10, LD50, LD90) with emphasis on the toxic effects that produce lethality in 10% of the animals, as this may represent a close approximation to the maximal tolerable dose in the species under study. NCI toxicology studies involve two phases. Acute toxicity evaluation first occurs in dose-range finding studies, which allow a bracketing of safe and likely toxic dose ranges. Second stage or more detailed IND-directed studies allow more definitive prediction of the probable safe starting dose and definition of the specific organ toxicity likely to be encountered in the patients. On reviewing experience using the initial human dose based on the LD10 data, this approach resulted in the risk of exceeding the maximum tolerable dose in humans of approximately 1%.¹⁶ Therefore, NCI generally selects the initial dose in humans as a fraction (eg 10%) of the LD10. DTP optimizes, schedules, develops pharmacological methods, conducts dose/range-finding studies, and performs IND-directed toxicity testing with correlative pharmacology and histopathology.

NCI Drug discovery for hematopoietic neoplasms

Since the inception of the 60 cell line assay over 75 000 compounds have been tested; approximately 7500 have been referred to for possible in vivo testing. Approximately 1400 of these have shown relative selectivity against the leukemia/lymphoma cell line panel. Of these, 80 compounds were further tested in the in vivo hollow fiber assay, and approximately 25% of these had activity. Certain of these include drugs in various stages of development including early phase clinical trials (Table 1). They represent structural classes that have not received prior clinical attention.

A substantial fraction of the screening results is available to the public through http://dtp.nci.nih.gov. For example, Table 2 contains a list of 66 'open' novel compounds, which have shown selective activity yet not fully delineated mechanism of action in vitro in leukemia/lymphoma cell lines. These compounds are available to the research community to further investigate and develop. The basis for this activity in relation to the presence and action of defined molecular targets in these cells will remain an opportunity for further research. Prominent examples of drugs detected by the screen that have advanced to the clinic include flavopiridol, UCN-01, and depsipeptide. Flavopiridol (NSC 649890) is a semi-synthetic flavone derived from rohitukine, an alkaloid isolated from a plant indigenous to India, Dysoxylum binectarieferum. In vitro cell culture and in vivo animal xenograft studies revealed that flavopiridol causes significant inhibition of various tumor cell lines including breast, lung, colon COL0205, prostate DU145, HL60 (promyelocytic leukemia cell line) and SUDHL4 (B cell lymphoma cell line), and head and neck HN-12.^17,18 Figure 3 graphically depicts the action in vitro of flavopiridol as it effectively displays the variations in sensitivity of various cell lines in the NCI screen to the agent. Initial mechanistic studies confirmed that flavopiridol represented a new class of chemotherapeutic agent, targeting cyclin-dependent kinases. Specifically, flavopiridol could induce cell cycle arrest during either G1 or G2/M, concordant with the ability to inhibit cdk2 and cdk1, respectively. It selectively inhibits cyclin-dependent kinases in comparison to other kinases. It is competitive with respect to ATP.¹⁹ Flavopiridol also down-regulates cyclin D1 and cyclin D3, important co-factors for cdk4 and cdk6 activation. Flavopiridol has also been shown to have antiangiogenic effect by down-regulating VEGF mRNA and limiting blood vessel formation in mouse matrigel model. Studies in several laboratories demonstrated that hemopoietic cell lines such as SUDHL-4 (B cell lymphoma), Jurkat (T cell leukemia/lymphoma) and CLL (chronic lymphocytic leukemia) cell lines were exquisitely sensitive to flavopiridol-induced apoptosis.^20,21,22,23 Mantle cell lymphoma cell lines that carry overexpression of cyclin D1 had induction of apoptosis and antiproliferative effect by flavopiridol.²⁴ These results have stimulated the development of clinical trials of flavopiridol in this specific refractory lymphoma. Two clinical trials of infusional flavopiridol administered a 72 h continuous infusion every 2 weeks.^25,26 The NCI phase I study of 72 h infusional flavopiridol treated 76 patients and documented one partial response in renal cancer, three minor responses in lymphoma, colon and renal cancers. Five patients were able to receive flavopiridol for more than a year and one patient for more than 2 years.²⁵ In the other phase I study, one patient with refractory metastatic gastric cancer achieved a sustained complete response without any evidence of disease more than 2 years after discontinuation of flavopiridol.²⁶ In vitro schedule-dependent synergy was found when flavopiridol followed treatment with paclitaxel, topotecan, doxorubicin and etoposide. This led to combination trials of flavopiridol with paclitaxel and cisplation resulting in clinical responses with esophageal and lung cancer.²⁷ Numerous phase II trials using a combination of flavopiridol with cytototoxic agent or as single agent in CLL (chronic lymphocyctic leukemia), non-Hodgkin's lymphoma, colon, gastric and head and neck cancer are underway at different clinical centers.

UCN-01, a derivative of staurosporine, originally isolated from Streptomyces species was found to be a nonspecific inhibitor of CDKs, PKC and causes cell arrest in G₁ and G₂ phases in different cell types. The phase I clinical trial did show evidence of activity in lymphomas,²⁸ with similar evidence reported in chronic lymphocytic leukemia (CLL) patients treated with a UCN-01 and fludarabine combination.²⁹ Depsipeptide derived from Chromobacterium violaceum, causes down-regulation of cyclin D1, up-regulation of cyclin E and p21-dependent cell cycle arrest at G₁ and G₂. Depsipeptide causes inhibition of histone acetylation and therefore suppression of transcriptional activity. Exposure to depsipeptide in vitro induced proapoptotic changes in B cell CLL including a decrease in bcl-2:bax ratio and p27 expression.³⁰ In a phase I study conducted at the NCI partial responses occurred in three patients with cutaneous T cell lymphoma and a complete response in a patient with peripheral T cell lymphoma.³¹ Hyperacetylation of histones was demonstrated in Sézary cells after treatment with depsipeptide in this study. A phase II study of depsipeptide in T cell lymphoma is now open for accrual at the NCI.

The completion of the human genome project and continuous discovery of newer targets and pathways in cell cycle regulation should lead to the identification and creation of compounds with specific functions and narrow toxicity profiles. CGAP (Cancer Genome anatomy project) is NCI's initiative to examine gene expression in different cancer types (http://cgap.nci.nih.gov). Particular genes can be correlated for their expression in drug-sensitive or -resistant cells using the micro array data on the website http://dtp.nci.nih.gov. The anticipated impact of the genome project and the molecular targeted approach to cancer treatment will not only be on novel therapies but also on strategies for diagnosis and prevention of malignancies.

Acknowledgements

We are indebted to Dr Richard Messmann for the figures.

1 Sausville EA, Feigal E. Evolving approaches to cancer drug discovery and development at the National Cancer Institute, USA. Ann Oncol 1999; 10: 1287-1291. MEDLINE

2 Sausville E, Shoemaker R. Role of the national cancer institute in acquired immunodeficiency syndrome-related drug discovery. J Natl Cancer Inst Monogr 2001; 28: 55-57. MEDLINE

3 Hodes L. Selection of molecular fragment features for structure-activity studies in antitumor screening. J Chem Inf Comput Sci 1981; 21: 132-136. MEDLINE

4 Driscoll JS. The pre-clinical new drug research program of the National Cancer Institute. Cancer Treat Rep 1984; 68: 63-76. MEDLINE

5 Alley MC, Scudiero DA, Monks A, Hursey ML, Czerwinski MJ, Fine DL. Feasibility of drug screening with panels of human tumor cell lines using a micro culture tetrazolium assay. Cancer Res 1988; 48: 589-601. MEDLINE

6 Monks A, Scudiero DA, Johnson GS, Paull KD, Sausville EA. The NCI anti-cancer drug screen: a smart screen to identify effectors of novel targets. Anticancer Drug Des 1997; 12: 533-541. MEDLINE

7 Hollingshead MG, Alley MC, Camalier RF, Abbott BJ, Mayo JG, Malspeis L. In vivo cultivation of tumor cells in hollow fibers. Life Sci 1995; 57: 131-141. MEDLINE

8 Johnson JI, Decker S, Zaharevitz D, Rubinstein LV, Venditti JM, Schepartz S, Kalyandrug S, Christian M, Arbuck S, Hollingshead M, Sausville EA. Relationships between drug activity in NCI preclinical in vitro and in vivo models and early clinical trials. Br J Cancer 2001; 84: 1424-1431. MEDLINE

9 Hall LA, Krauthauser CM, Wexler RS, Hollingshead MG, Slee AM, Kerr JS. The hollow fiber assay: continued characterization with novel approaches. Anticancer Res 2000; 20: 903-911. MEDLINE

10 Paull KD, Shoemaker RH, Hodes L, Monks A, Scudiero DA, Rubinstein L et al. Display and analysis of patterns of differential activity of drugs against human tumor cell lines: development of mean graph and COMPARE algorithm. J Natl Cancer Inst 1989; 81: 1088-1092. MEDLINE

11 Alvarez M, Paull K, Monks A, Hose C, Lee JS, Weinstein J, Grever M, Bates S, Fojo T. Generation of a drug resistance profile by quantitation of mdr-1/P-glycoprotein in the cell lines of the National Cancer Institute Anticancer Drug Screen. J Clin Invest 1995; 95: 2205-2214. MEDLINE

12 Weinstein JN, Myers TG, O'Connor PM, Friend SH, Fornace AJ Jr, Kohn KW, Fojo T, Bates SE, Rubinstein LV, Anderson NL, Buolamwini JK, van Osdol WW, Monks AP, Scudiero DA, Sausville EA, Zaharevitz DW, Bunow B, Viswanadhan VN, Johnson GS, Wittes RE, Paull KD. An information-intensive approach to the molecular pharmacology of cancer. Science 1997; 275: 343-349. Article MEDLINE

13 Scherf U, Ross DT, Waltham M, Smith LH, Lee JK, Tanabe L, Kohn KW, Reinhold WC, Myers TG, Andrews DT, Scudiero DA, Eisen MB, Sausville EA, Pommier Y, Botstein D, Brown PO, Weinstein JN. A gene expression database for the molecular pharmacology of cancer. Nat Genet 2000; 24: 236-244. Article MEDLINE

14 Koo HM, Monks A, Mikheev A, Rubinstein LV, Gray-Goodrich M, McWilliams MJ. Enhanced sensitivity to 1-beta-D-arabinofuranosylcytosine and topoisomerase II inhibitors in tumor cell lines harboring activated ras oncogenes. Cancer Res 1996; 56: 5211-5216. MEDLINE

15 Neubauer A, Dodge RK, George SL, Davey FR, Silver RT, Schiffer CA, Mayer RJ, Ball ED, Wurster-Hill D, Bloomfield CD, Liu ED. Prognostic importance of mutations in the ras proto-oncogenes in de novo acute myeloid leukemia. Blood 1994; 83: 1603-1611. MEDLINE

16 Schein P, Anderson T. The efficacy of animal studies in predicting clinical toxicity of cancer chemotherapeutic drugs. Int J Clin Pharmacol 1973; 8: 228-238. MEDLINE

17 Patel V, Senderowicz AM, Pinto D Jr, Igishi T, Raffeld M, Quintanilla-Martinez L. Flavopiridol, a novel cyclin-dependent kinase inhibitor, suppresses the growth of head and neck squamous cell carcinomas by inducing apoptosis. J Clin Invest 1998; 102: 1674-1681. MEDLINE

18 Arguello F, Alexander M, Sterry JA, Tudor G, Smith EM, Kalavar NT. Flavopiridol induces apoptosis of normal lymphoid cells, causes immunosuppression, and has potent antitumor activity in vivo against human leukemia and lymphoma xenografts. Blood 1998; 91: 2482-2490. MEDLINE

19 Carlson BA, Dubay MM, Sausville EA, Brizuela L, Worland PJ. Flavopiridol induces G1 arrest with inhibition of cyclin-dependent kinase (CDK) 2 and CDK4 in human breast carcinoma cells. Cancer Res 1996; 56: 2973-2978. MEDLINE

20 Arguello F, Alexander M, Sterry J, Yudor G, Smith E, Kalavar N, Greene J, Koss W, Morgan D, Stinson S, Siford T, Alvord W, Labansky R, Sausville E. Flavopiridol induces apoptosis of normal lymphoid cells, causes immunosuppression, and has potent antitumor activity in vivo against human leukemia and lymphoma xenografts. Blood 1998; 91: 2482-2490. MEDLINE

21 Byrd JC, Shinn C, Waselenko JK, Fuchs EJ, Lehman TA, Nguyen PL, Flinn IW, Diehl LF, Sausville EA, Grever MR. Flavopiridol induces apoptosis in chronic lymphocytic leukemia cells via activation of caspase-3 without evidence of bcl-2 modulation or dependence on functional p53. Blood 1998; 92: 3804-3816. MEDLINE

22 Parker BW, Kaur G, Nieves-Neira W, Taimi M, Kohlhagen G, Shimizu T, Losiewicz MD, Pommier Y, Sausville EA, Senderowicz AM. Early induction of apoptosis in hematopoietic cell lines after exposure to flavopiridol. Blood 1998; 91: 458-465. MEDLINE

23 Kitada S, Zapata JM, Andreeff M, Reed JC. Protein kinase inhibitors flavopiridol and 7-hydroxy-staurosporine down- regulate antiapoptosis proteins in B-cell chronic lymphocytic leukemia. Blood 2000; 96: 393-397. MEDLINE

24 Guedez L, Quintanilla-Martinez L, Lahusen T, Davies T, Singh SS, Barotto N, Vistica D, Raffeld M, Sausville EA, Senderwicz AM. Flavopiridol-induced apoptosis is associated with a decrease in cyclin D1 in mantle lymphoma cell lines Proc 90th Annual Meeting of the American Association of Cancer Research, Philadelphia, PA, 1999.

25 Senderowicz AM, Headlee D, Stinson SF, Lush RM, Kalil N, Villalba L, Hill K, Steinberg SM, Figg WD, Tompkins A, Arbuck SG, Sausville EA. Phase I trial of continuous infusion flavopiridol, a novel cyclin-dependent kinase inhibitor, in patients with refractory neoplasms. J Clin Oncol 1998; 16: 2986-2999. MEDLINE

26 Thomas J, Cleary J, Tutsch K, Arzoomanian R, Alberti D, Simon K, Feierabend D, Morgan K, Wilding G. Phase I clinical and pharmokinetic trial of flavopiridol Proc 88th Annual Meeting of the American Association of Cancer Research, San Diego, CA, 1997.

27 Shah MA, Kaubisch A, O'Reilly S. A phase IB clinical trial of the sequence dependent combination of paclitaxel (P) and cisplatin (C) with cyclin dependent kinase (CDK) inhibitor flavopiridol in patients with advanced solid tumors. Proceedings of AACR 2001; 42: 2917a.

28 Sausville EA, Arbuck SG, Messmann R, Headlee D, Bauer KS, Lush RM. Phase I trial of 72-hour continuous infusion UCN-01 in patients with refractory neoplasms. J Clin Oncol 2001; 19: 2319-2333. MEDLINE

29 Wilson WH, Gutierrez M, Stetler-Stevenson M, Drbohlav N, Staudt L, Figg W, Sausville EA, Fowler D, Bishop M, Hegde U. Phase I trial of 7-hydroxystaurosporine (UCN-01) and fludarabine phosphate; in vivo evidence of UCN-01 induced apoptosis in CLL. ASH Proc 2000; 96: 3268a.

30 Byrd JC, Shinn C, Ravi R, Willis CR, Waselenko JK, Flinn I, Dawson NA, Grever MR. Depsipeptide (FR901228): a novel therapeutic agent with selective, in vitro activity against human B-cell chronic lymphocytic leukemia cells. Blood 1999; 94: 1401-1408. MEDLINE

31 Bates SE, Sandor V, Bakke S. A phase I study of FR901228 (Depsipeptide), a histone deacetylase inhibitor. ASCO Proc 1999; 18: 693a.

Figure 1 Developmental Therapeutics Program (DTP) organization structure consisting of nine branches and one clinical unit.

Figure 2 Schema for compound screening at the National Cancer Institute's Developmental Therapeutics program (DTP).

Figure 3 The mean graph representation of antitumor effects of flavopiridol in the in vitro cancer screen. The data summarized present a visual image consistent with results obtained from the NCI cancer screen. The individual response of each cell line to the agent is depicted by a bar graph extending either to the right or left of the mean, depending on whether the cell line was either more or less sensitive than the average response. The length of each bar is proportional to the relative sensitivity compared with the mean determination. A characteristic 'fingerprint' of cellular responsiveness can be visualized from the data and identify a known or yet unidentified molecular target. The data show clearly that flavopiridol is highly active in the HL60, MOLT-4 and SR hematopoietc neoplasm cell lines.

Table 1 List of compounds recognized by unique performance initially in the in vitro drug screen in various stages of development by the Developmental Therapeutics Program at the National Cancer Institute

Table 2 List of 'open' compounds with unidentified mechanism of actions in the NCI in vitro screen, but selective activity in hematopoietic neoplasm-derived cell lines