Phase i cancer trials


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Phase i cancer trials

  1. 1. ASCO SPECIAL ARTICLE Critical Role of Phase I Clinical Trials in Cancer Treatment Adopted on November 8, 1996 by the American Society of Clinical Oncology*P HASE I clinical trials are the essential gateways to the development of new cancer therapies. The pro-cess of translating basic research findings into clinical the murine LD10 , although it may be even lower if other species (ie, dogs) were more sensitive to the drug. Extensive preclinical testing is required prior to initiat-applications necessarily commences in small phase I trials ing the first phase I study. The candidate drugs may eitherthat lead to larger investigations and eventual regulatory be synthesized, which often results in a group of analogsapproval and widespread usage. Despite the centrality of to choose from, or identified by screening extracts ofthese early-phase trials to the process of medical discov- natural products. Then, a lead candidate is selected basedery, they are not well understood and are indeed subject on its relative activity in a variety of experimental tumorto significant misconceptions, particularly as they relate models. These models may include tumor cells that areto clinical oncology research.1 grown in tissue culture and/or animals who are bearing Traditionally, phase I trials have involved the adminis- tumors. Subsequent studies generally will include preclin-tration of usually subtherapeutic doses of a new agent to ical efficacy studies, which show beneficial activity, andhealthy volunteers to assess toxicity. In contrast, phase I preclinical toxicology studies, which suggest that the drugcancer trials can represent a real therapeutic option for will likely be safe at effective doses. As noted previously,some patients who have failed to respond to other treat- such studies also determine the human starting dose. Inment or for whom no other therapies exist. Because of addition, preclinical pharmacology studies often providethe importance of phase I trials for both research and extensive information about pharmacokinetics.treatment purposes, the American Society of Clinical On- It is common for drug discovery and development tocology (ASCO) convened a subcommittee of the Public be directed based on experience with another drug. TheIssues Committee to review current issues that are related recent development and approval of two camptothecinto these trials and to draft a policy statement regarding analogs, topotecan and irinotecan, exemplify such di-their role in cancer treatment and research. rected development. Camptothecin was developed more than 20 years ago, but was abandoned because of severe WHAT ARE PHASE I CLINICAL TRIALS? bladder toxicity, diarrhea, and myelosuppression.4 Be- Phase I trials classically are considered ‘‘first in hu- cause of its marked preclinical activity, there was a longman’’ studies. For most drugs outside oncology, such search for acceptable analogs, which culminated in thephase I studies are conducted in healthy volunteers in approval this year by the Food and Drug Administrationspecially dedicated clinical pharmacology units. How- (FDA) of topotecan and irinotecan for refractory ovarianever, because of the toxicity that generally is observed in and colorectal cancers, respectively.preclinical studies, phase I studies of new anticancer In the United States, there may be extensive regulatoryagents almost always are conducted in patients with re-fractory cancers. Most phase I studies are cohort studies, in which pa- See Appendix for Subcommittee on Phase I Clinical Trials membertients are treated at increasing doses according to chrono- affiliations.logical entry into the study.1,2 Thus, results in early pa- From the American Society of Clinical Oncology.tients greatly influence dosing of subsequent patients. The Submitted November 27, 1996; accepted November 27, 1996. Address reprint requests to American Society of Clinical Oncol-starting dose is based on preclinical testing, and is usually ogy, 225 Reinekers Lane, Suite 650, Alexandria, VA 22314.quite conservative. A standard measure of toxicity of a *ACSO sincerely appreciates the contributions of the ASCO Sub-drug in preclinical testing is the percentage of animals committee on Phase I Clinical Trials, which devoted much time and(rodents) who die because of treatment. The dose at which effort to this project. Chair: Mark J. Ratain; Subcommittee Mem-10% of the animals die is known as the LD10 , which has bers: Deborah Collyar, Barton A. Kamen, Elizabeth Eisenhauer, Theodore Steven Lawrence, Carolyn Runowicz, Sam Turner, andin the past often correlated with the maximal-tolerated James L. Wade III.dose (MTD) in humans, adjusted for body-surface area.3 ᭧ 1997 by American Society of Clinical Oncology.Thus, the standard conservative starting dose is one tenth 0732-183X/97/1502-0055$3.00/0Journal of Clinical Oncology, Vol 15, No 2 (February), 1997: pp 853-859 853AID JCO 0071 / 5f1a$$1401 01-28-97 18:34:51 jcoa WBS: JCO
  2. 2. 854 AMERICAN SOCIETY OF CLINICAL ONCOLOGYreview prior to the first phase I study. The sponsor must about these trials. It is commonly misstated that suchfile an Investigational New Drug (IND) application, to trials are nontherapeutic toxicology studies, that phase Iinclude all the preclinical data and the initial clinical studies pose high risk of extreme toxicity, that cancerprotocol(s). The FDA has 30 days in which to decide patients are too vulnerable to give informed consent, andwhether such protocols may proceed without additional that the sponsor of the drug covers all costs of suchpreclinical data or modification of the clinical protocol. studies.There are other levels of review, which often occur priorto submission to the FDA. Most large pharmaceutical Misconception: Phase I Trials Are Not Therapeuticcompanies have their own internal protocol review com- As noted previously, phase I studies of most drugsmittees that examine the protocols for both ethical and are conducted in healthy volunteers. However, almost allscientific issues. Compounds under development by the phase I studies of new anticancer agents are conducted inNational Cancer Institute (NCI) have extensive review patients with malignancies that are refractory to standardthrough the NCI Decision Network, and each protocol therapy or for which there is no standard therapy. Theundergoes detailed ethical and scientific scrutiny prior to development of a new agent has traditionally consistedFDA filing. In addition, there is extensive review at the of its sequential study in trials that have different majorinstitutional level that must include approval by an Insti- endpoints. Although the goals for each phase of develop-tutional Review Board, and often a cancer center scientific ment may shift, the hope in the treatment of individualreview committee also. patients on such trials is that the new therapy will offer It is important to recognize that phase I studies are not therapeutic benefit. These subjects are patients who arelimited to ‘‘first in human’’ studies. Subsequent phase I seeking palliation of their disease and who recognize thestudies often evaluate new schedules or combinations investigational nature of phase I studies. It should not bewith established drugs or radiation. In addition, these sec- relevant whether the scientific goals of the study are toondary phase I studies may evaluate toxicity and pharma-cokinetics in patient populations that were excluded in determine toxicity and pharmacokinetics (phase I) or toprior studies, such as children. ascertain the response rate (phase II) because the patient is Phase I studies represent the critical transition point receiving an appropriate treatment for his or her disease.from the laboratory to future improvements in cancer care Phase I trials represent the first translational step fromand outcomes. All drugs that are currently marketed first the laboratory into patients. The decision to move anshowed activity in phase I studies.5 Two of today’s most agent into phase I evaluation is based on extensive pre-commonly prescribed drugs, cisplatin and paclitaxel, ex- clinical evaluation, as detailed previously. The centralemplify these successes. criterion is the observation of sufficient preclinical antitu- Cisplatin was best known for its toxicity during its mor activity, such that a therapeutic effect in human can-initial clinical evaluation. It induced severe nausea and cer is anticipated. Thus, although the goal of a phase Ivomiting in an era of only modestly effective antiemetics. study is to identify the recommended dose for futureIt induced acute renal failure prior to the development of trials, there is reasonable expectation that antitumor ef-hydration strategies to prevent such complications. And fects will be noted in some patients in the induced major responses in patients with chemotherapy- There is a tension that is inherent in phase I designresistant testes cancer.6 In the absence of these responses, between the need to balance the concern for patient safetycisplatin probably would have been abandoned. when being treated with an unknown agent, as reflected Paclitaxel, then known as taxol, also was an extremely in careful dose escalation, and the desire to treat at dosesdifficult agent for its phase I investigators. The major that will be close to the recommended phase II dose, thustoxicity to be reckoned with was hypersensitivity reac- increasing the likelihood of benefit. This has led to ations, which since have been improved by pretreatment number of proposals that permit more rapid dose escala-with corticosteroids and antihistamines. Again, there was tion, some of which investigators are now implement-great enthusiasm for further development, despite its tox- ing.1,2,8 Unfortunately, translating animal studies to hu-icity, because of responses observed in women with re- mans is sometimes difficult, such that the necessity offractory ovarian cancer.7 starting low is appropriate. Accelerating the dose escala- tion through pharmacodynamically driven studies may COMMON MISCONCEPTIONS REGARDING help to optimize therapy. PHASE I TRIALS The usual measurement of therapeutic benefit during Because phase I studies are unfamiliar to most physi- phase I trials has been the assessment of objective tumorcians and patients, there are many popular misconceptions regression (response). Evidence from the literature sug-AID JCO 0071 / 5f1a$$1402 01-28-97 18:34:51 jcoa WBS: JCO
  3. 3. CRITICAL ROLE OF PHASE I CLINICAL TRIALS 855gests that, although response rates in phase I trials can gation of the infusion, could reduce toxicity and maintainbe low, they very often are helpful in identifying which activity.16agents subsequently will be of benefit and in directingsubsequent phase II investigations.5 In fact, it has been Misconception: Phase I Trials in Patients withsuggested that failure to observe responses in phase I Advanced Cancer Are Ethically Questionabletrials is predictive of subsequent failure of the agent and Patients who are offered treatment in the context of ashould be considered in the decision of moving the agent phase I trial are informed that the purpose of the studyinto phase II development.9 is to find the optimal drug dose and that some patients Aside from response, other measures of therapeutic will be treated at a dose that is too low or too high (abovebenefit may be documented in clinical trials. These in- the MTD). Investigators also inform them that there isclude improvement in disease-related symptoms, tumor a possibility, although small, that the treatment will bemarkers, and global quality of life. Unfortunately, little beneficial, as important new drugs have always been ac-is known about the success of phase I agents in inducing tive in phase I testing in refractory cancers. The Institu-changes in these measures.10 Nevertheless, the growing tional Review Board, the sponsor, and the FDA (wheninterest in endpoints in clinical trials that overall are re- applicable) approve the written consent forms.lated to quality of life may have an impact on phase I The misconception seems to center around the beliefstudies. that phase I trials are ethically questionable because of In summary, although the goal of a phase I trial is to the ‘‘vulnerable’’ population that is considered in thesedetermine the toxic effects, pharmacologic behavior, and trials.17 These patients are ‘‘vulnerable’’ to the extent thatrecommended doses for future study of a new agent, there they know they are going to die from their disease. Withis a strong preclinical rationale for bringing the drug into that knowledge, and following an informed discussion,the clinic with the expectation of positive clinical out- they may then consider which course of action they maycomes for some patients. In fact, Institutional Review want to take. Such action may be no treatment, treatmentBoards would not permit the administration of potentially with chemotherapy (or other conventional approaches)toxic treatments to patients unless there was some reason- outside of a clinical trial, treatment with unorthodox ther-able prospect of antitumor effect.11 apies, or participation in a clinical trial. Many of them will decide not to enroll in a trial, but some will. And such patients who participate in phase I trials appear toMisconception: Patients Participating in Phase I Trials have adequate (self-perceived) knowledge of the risks ofMust Expect Severe Toxicity investigational agents.18 The only time these patients may A major endpoint of phase I trials of a new agent is the be considered incapable of analyzing their risks is whendefinition of the MTD that causes dose-limiting toxicity their death is imminent. At this point in their illness, they(DLT). As described previously, the trials usually begin would certainly not be eligible for any clinical a conservative dose that is expected to cause no toxic-ity. In the classic design, it is expected that there will be Misconception: All Costs of Phase I Trials Arefour to six escalations of dose before reaching the MTD. Covered by SponsorTherefore, the subsequent phase II trial of an agent usu- Third-party payors seek to justify denial of coverageally results in more toxicity than the phase I because more for routine patient care costs in clinical trials, perhapspatients will be treated at the presumed MTD.1 especially phase I trials, on the ground that such costs Toxic deaths especially are rare in phase I trials, oc- are the responsibility of the pharmaceutical sponsor orcurring in only 0.5% of adult patients.12 Similar results other entity who is conducting the research. This is ahave been shown in children.13 Phase I trials have not widespread misconception that reflects the views of third-shown an adverse impact on either patients’ quality of party payors, but has no basis in historical fact. For aslife or survival.10,14 long as clinical research has been conducted, payment for Sometimes, second-generation phase I trials of particu- routine patient care costs, such as physician and hospitallarly promising agents specifically may aim to develop charges, has been borne by insurers or individual patients.approaches to circumvent toxicities. This particularly was One exception to this rule is the Clinical Center of thecritical for the development of paclitaxel, which engen- National Institutes of Health, where an explicit policy ofdered severe hypersensitivity reactions in initial trials.15 providing patient care free of charge has always been inSubsequent phase I studies showed that premedication effect. However, even the Clinical Center is now underwith corticosteroids and antihistamines, as well as prolon- pressure to collect third-party payment where it is avail-AID JCO 0071 / 5f1a$$1402 01-28-97 18:34:51 jcoa WBS: JCO
  4. 4. 856 AMERICAN SOCIETY OF CLINICAL ONCOLOGYable, either through private insurance or public programs help guide the selection of tumor types for subsequentsuch as Medicare. phase II and III testing. Although third-party payors customarily cover routinepatient care costs, the research sponsor is responsible for Patients May Be Treated With Relatively Ineffectivethe costs of the investigational agent and its incremental and/or More Toxic Therapiescosts, and of data collection, management, and analysis. Systemic chemotherapy for most metastatic solid tu-Responsibility for patient-care costs has become an item mors provides only temporary palliation. When diseaseof contention only over the past decade, when sometimes progression occurs during treatment, patients and theirexpensive new therapies (such as high-dose chemother- doctors have three options: continue the current ineffec-apy and autologous bone marrow transplant for breast tive regimen; stop systemic treatment altogether; or try acancer) administered in a research setting have attracted different regimen. Many patients who have a good perfor-the attention of claims reviewers and resulted in reim- mance status will select the third option. Although somebursement denials.19 Undue focus on bone marrow trans- patients may benefit from second-line or even third-lineplant, with its atypical costs, has distorted the perception treatment regimens, there are some chemotherapeuticof third-party payors about clinical research in general. drugs that are used in this setting that have no measurable Third-party payors also take the position that they have clinical benefit. One example of such a treatment is theno contractual obligation to cover treatment that is given use of mitomycin for patients with colorectal a clinical trial because such treatment is ‘‘experimen- Some clinical trials of this drug for this disease havetal’’ or ‘‘investigational.’’ Such experimental exclusion measured a response rate of 0%.20 Therefore, it is clearclauses originally were inserted to prevent patients from that, in at least some common clinical situations, phase Ibeing subjected to unorthodox approaches to serious ill- trials have a greater probability of benefit than nonexperi-ness, and were intended for the patient’s protection. Now, mental options.such exclusions are used to prevent patients from receiv-ing therapies that are recommended by leading physi- Patients May Be Treated With More Costly Therapiescians, thereby protecting payors’ finances but placing pa- Patients who have failed all standard therapies and whotients’ health at risk. In fact, for many cancer patients wish to enroll in, but who are denied access to, a phasewho have failed other therapy, clinical trials, especially I trial continue to require medical care. Such patientsphase I trials, may represent not only the best, but one may receive treatment with an alternative commerciallyof the few justified therapeutic treatment options. In that available regimen or supportive care, either of which maycontext, if insured patients are to receive the value of be more expensive than the therapy in the denied phasetheir insurance policy or their eligibility under public I trial.21programs such as Medicare, exclusions for experimentaltreatment should not be automatically applied. Investiga- Decreased Pharmacologic Knowledge Abouttional therapy is widely accepted as part of the standard Anticancer Drugsof care in oncology, and insurance plans that fail to recog- Phase I studies provide the best opportunity to elucidatenize this fact are depriving their policyholders and bene- the often complex clinical pharmacology of a potentiallyficiaries of significant treatment alternatives. effective agent. These studies generally involve adminis- tering a broad range of doses, which allows elucidation CONSEQUENCES OF MISCONCEPTIONS of dose-toxicity relationships. Furthermore, issues, such as drug excretion and metabolism, can be analyzed viaDecreased Availability of New Drugs for Phase II/III detailed studies in small numbers of patients in a singleDevelopment institution. After completion of phase I, studies often are Because the progress of new agents into phase II and conducted at multiple sites, which complicates the collec-III has, as an absolute requirement, the successful comple- tion and analysis of pharmacologic data. Further studiestion of (usually) more than one phase I trial, it follows often are conducted in combination with establishedlogically that a restriction in access to phase I agents by agents, which makes assessment of information about thewhatever means inevitably will result in a decrease in the new agent especially difficult. Any decrease in the num-availability of agents for phase II and III development. ber of phase I trials and, as a consequence, the amount ofFurthermore, a restriction on the number of phase I trials pharmacologic data accumulated, may hinder subsequentcarried out with any individual drug may limit the oppor- rational development and safe use of the drug.tunities to observe evidence of therapeutic activity, which Advances in laboratory techniques have facilitatedAID JCO 0071 / 5f1a$$1402 01-28-97 18:34:51 jcoa WBS: JCO
  5. 5. CRITICAL ROLE OF PHASE I CLINICAL TRIALS 857measurement of plasma concentrations of potent antican- ticularly acute with respect to early phase trials, whichcer drugs, which has led to an improvement in their use.22 are shrouded in misconceptions about both toxicity andMoreover, our understanding of the genetics of drug- potential therapeutic benefit. The goal of the clinical re-metabolizing enzymes is creating new paradigms in treat- search community should be to enhance public awarenessment.23 It may be possible to test a patient for a particular and understanding about the pivotal role of phase I trialsdrug-metabolizing gene prior to treatment and then pre- in the overall research enterprise and in the comprehen-scribe an individualized dose, rather than our current ap- sive treatment of people with cancer. Fulfillment of thisproach of prescribing an average dose to everybody. Such goal would lead to greater research funding, a more effi-an approach has been shown to be feasible.24 cient drug-development process, and more widespread acceptance of early phase trials as valid treatment options.Curtailed Development to Minimize Number of Phase ITrials Increased Access to Phase I Trials as a Legitimate All phase I trials have limitations on patient eligibility. Therapeutic Option in the Context of Managed CareSome of these limitations are in the interest of patient The impact of managed care on clinical research is ofsafety, whereas others are included to ensure a relatively substantial concern in the oncology community. Patientshomogeneous patient population. As a consequence, and physicians alike believe that decision-making bysome patient populations are excluded and optimal dosing managed care organizations may be driven more by con-in these patients may be unknown, even after full FDA siderations of cost than in the fee-for-service system.approval. Specific populations include children, the very Moreover, those who have organized managed care plansold, and patients with marked organ dysfunction second- often do not seem to place a great premium on the capac-ary to either the disease or complications of prior therapy. ity to conduct clinical research.One approach to resolving this is additional phase I trials With greater education and public awareness of thein special populations. If phase I trials were more costly value of clinical research, this situation may be sponsors (because of no reimbursement for patient care Managed care plans will eventually come to compete notcosts), such trials would not be performed. Instead, pa- merely on cost, but also on quality of care. If potentialtients in these special populations would have to be enrollees and other purchasers of managed care productstreated based on the experience of their oncologist, rather are convinced that quality cancer care requires access tothan on the basis of scientific data. clinical trials, including phase I trials, access to investiga- Another area in which phase I trials are helpful but not tional therapy may become a marker for quality in com-always pursued is in the development of new schedules parison of different plans. In addition, cost-consciousor new combinations. Such approaches can be applied plans will value the data that are made available to themdirectly in phase II trials, but if the doses are incorrect, by their participation in clinical research.such trials may be useless (if at too low a dose) or lethal(if at too high a dose). Specific phase I trials that address Reimbursement for All Patient Care Costs Associatedthese issues can rapidly provide reliable data, but are with Phase I Clinical Trialsunlikely to be performed if the costs of such studies aretoo high. Beginning with the health care reform debate of 1993- 1994, a number of legislative proposals have been intro- PROPOSED IMPROVEMENT IN PROCESS duced in the United States Congress to require coverage (AND RAMIFICATIONS) of patient care costs in clinical trials. An important feature of all these proposals is that they fail to distinguish amongIncreased Public Awareness Regarding Therapeutic different phases of clinical research. Instead, the propos-and Societal Value of Drug Development als recognize that, in appropriate circumstances, all phases There is a widespread misunderstanding among the of clinical research may represent treatment options forpublic concerning the nature of clinical trials, and more people with life-threatening diseases such as cancer. Al-specifically, the manner in which patients are able to though these proposals have yet to become law, they areparticipate in clinical cancer research for treatment pur- redefining the manner in which third-party payors in bothposes. A fully informed public would support clinical private and public plans are evaluating their coveragecancer research, not just for the purpose of advancing policies. If sympathetic legislators continue to place theknowledge about cancer, but also to provide optimal treat- question of reimbursement in a research setting beforement opportunities. The lack of public knowledge is par- the American people, all phases of high-quality peer-AID JCO 0071 / 5f1a$$1402 01-28-97 18:34:51 jcoa WBS: JCO
  6. 6. 858 AMERICAN SOCIETY OF CLINICAL ONCOLOGYreviewed clinical research one day may be available to exists.25 These efforts can be enhanced by targeting grantpeople with cancer and other life-threatening diseases. support to this important area. SUMMARYResearch Regarding Phase I Trial Designs That The physician/investigator must simultaneously man-Maximize Likelihood Of Patient Benefit age the twin goals of patient benefit and knowledge acqui- There has been only limited research into design of sition. There are few fields with greater challenges thanphase I trials, and most published studies have used grad- early clinical trials of anticancer agents, as all clinicalual dose-escalation schemes with three to six patients per protocols must be both scientifically and therapeuticallycohort. However, in the past 5 years, there have been a valid. The task of the physician/investigator would benumber of published suggestions for modification of the greatly facilitated if research is performed in an environ-standard trial paradigm.1,2,8 In the ideal phase I study, ment in which the public, which includes third-party pay-dose escalation will rapidly proceed to doses near the ors, is adequately informed and sufficiently supportive ofMTD, without drug-related fatalities. This will provide the role of clinical research in the treatment of cancer.the maximal chance of therapeutic benefit to the subjects Only through sustained and even enhanced support ofthat are willing to take on the risks of such trials.5 It also early-phase testing will the health care system take fullappears feasible to empower patients to participate in advantage of the many new basic science discoveries thatthe selection of their dose, where reasonable uncertainty are awaiting translation into clinical application. APPENDIX Subcommittee on Phase I Clinical Trials (As of 6/28/96) Investigator Institution Mark J. Ratain, MD University of Chicago, Chicago, IL Deborah Collyar, PhD, RN Patients Advocates in Research, and Clinical Trials Information Project, Black Hawk, CA Barton A. Kamen, MD, PhD University of Texas Southwestern Medical Center, Dallas, TX Elizabeth Eisenhauer, MD Queen’s University, Kingston, Ontario, Canada Theodore Steven Lawrence, MD University of Michigan Medical Center, Ann Arbor, MI Carolyn Runowicz, MD Albert Einstein College of Medicine, Bronx, NY Sam Turner Fox, Bennett & Turner, Washington, DC James L. Wade III, MD Cancer Care Specialists of Central Illinois, Decatur, IL REFERENCES 1. Ratain MJ, Mick R, Schilsky RL, et al: Statistical and ethical for the future clinical activity of a new agent? Ann Oncol 7:14, 1996issues in the design and conduct of phase I and II clinical trials of (abstr; suppl 1)new anticancer agents. J Natl Cancer Inst 85:1637-1643, 1993 10. Melink TJ, Clark GM, Von Hoff DD: The impact of phase I 2. Arbuck SG: Workshop on phase I study designs—Ninth NCI/ clinical trials on the quality of life in patients with cancer. AnticancerEORTC New Drug Development Symposium, Amsterdam, March Drugs 3:571-576, 199212, 1996. Ann Oncol 7:567-573, 1996 11. Kodish E, Stocking C, Ratain MJ, et al: Ethical issues in 3. Freireich EJ, Gehan EA, Rall DP, et al: Quantitative compari- phase I oncology research: A comparison of investigators and IRBson of toxicity of anticancer agents in mouse, rat, hamster, dog, chairpersons. J Clin Oncol 10:1810-1816, 1992monkey, and man. Cancer Chemother Rep 50:219-244, 1966 12. Decoster G, Stein G, Holdener EE: Responses and toxic 4. Wall ME, Wani MC: Camptothecin and taxol: Discovery to deaths in phase I clinical trials. Ann Oncol 1:175-181, 1990clinic—Thirteenth Bruce F. Cain Memorial Award Lecture. Cancer 13. Weitman S, Ochoa S, Sullivan N, et al: Pediatric phase IIRes 55:753-760, 1995 cancer chemotherapy trials: A Pediatric Oncology Group Study. Proc 5. Von Hoff DD, Turner J: Response rates, duration of response, Am Soc Clin Onc 15:463, 1996 (abstr 1456)and dose response effects in phase I studies of antineoplastics. Invest 14. Janisch L, Mick R, Schilsky RL, et al: Prognostic factors forNew Drugs 9:115-122, 1991 survival in patients treated on phase I clinical trials. Cancer 74:1965- 6. Rossof AH, Slayton RE, Perlia CP: Preliminary clinical experi- 1973, 1994ence with cisdiamminedichloroplatinum (II) (NSC 119875, CACP). 15. Weiss RB, Donehower RC, Wiernik PH, et al: Hypersensitiv-Cancer 30:1451-1456, 1972 ity reactions from taxol. J Clin Oncol 8:1263-1268, 1990 7. Donehower RC, Rowinsky EK, Grochow LB, et al: Phase I 16. Wiernik PH, Schwartz EL, Einzig A, et al: Phase I trial oftrial of taxol in patients with advanced cancer. Cancer Treat Rep taxol given as a 24-hour infusion every 21 days: Responses observed71:1171-1177, 1987 in metastatic melanoma. J Clin Oncol 5:1232-1239, 1987 8. Dent SF, Eisenhauer EA: Phase I trial design: Are new method- 17. Annas GJ: The changing landscape of human experimenta-ologies being put into practice? Ann Oncol 7:561-566, 1996 tion: Nuremberg, Helsinki, and beyond. Health Matrix 2:119-140, 9. Von Hoff DD: Is response in phase I trials a useful predictor 1992AID JCO 0071 / 5f1a$$1402 01-28-97 18:34:51 jcoa WBS: JCO
  7. 7. CRITICAL ROLE OF PHASE I CLINICAL TRIALS 859 18. Daugherty C, Ratain MJ, Grochowski E, et al: Perceptions validation of a pharmacologically based dosing scheme for the cis-of cancer patients and their physicians involved in phase I trials. J diamminechloroplatinum(II) analogue diamminecyclobutanedicar-Clin Oncol 13:1062-1072, 1995 boxylatoplatinum. Cancer Res 45:6502-6506, 1985 19. Peters WP, Rogers MC: Variation in approval by insurance 23. Lu Z, Diasio RB: Polymorphic drug-metabolizing enzymes,companies of coverage for autologous bone marrow transplantation in Schilsky RL, Milano GA, Ratain MJ (eds): Principles of Antineo-for breast cancer. N Engl J Med 330:473-477, 1994 plastic Drug Development and Pharmacology. New York, NY, Mar- 20. Poplin EA, LoRusso P, Lokich JJ, et al: Randomized clinical cel Dekker, 1996, pp 281-305trial of mitomycin-C with or without pretreatment with WR-2721 24. Ratain MJ, Mick R, Janisch L, et al: Individualized dosingin patient with advanced colorectal cancer. Cancer Chemother Phar- of amonafide based on a pharmacodynamic model incorporatingmacol 33:415-419, 1994 acetylator phenotype and gender. Pharmacogenetics 6:93-101, 21. Jaakkimainen L, Goodwin PJ, Pater J, et al: Counting the 1996costs of chemotherapy in a national cancer institute of Canada ran- 25. Daugherty C, Lyman K, Minami H, et al: A feasibility studydomized trial in nonsmall-cell lung cancer. J Clin Oncol 8:1301- of informed consent and medical decision-making employing an1309, 1990 interactive patient choice design in phase I trials. Proc Am Soc Clin 22. Egorin MJ, Van Echo DA, Olman EA, et al: Prospective Oncol 15:175, 1996 (abstr)AID JCO 0071 / 5f1a$$1404 01-28-97 18:34:51 jcoa WBS: JCO