| Jump to: |
| GENERAL INFORMATION (Back to top) Many new cancer therapies become available to patients through clinical trials. Treatment studies that involve drugs or invasive procedures are categorized by phase. Phase I studies generally establish whether a treatment is safe and at what dosages. Phase II studies assess the efficacy of treatments after their safety and feasibility has been established in Phase I studies. Phase III studies compare effective treatments from phase II studies to currently accepted treatments. Phase IV studies collect and compare data on established treatments. Studies can also be categorized as adjuvant and neoadjuvant. Adjuvant studies evaluate treatments that try to prevent the recurrence of cancer after a patient has become clinically free of disease. Neoadjuvant studies evaluate treatments designed to reduce the extent of tumor to a point where it can be definitively treated by therapies that are considered standard. Besides evaluating cancer treatment, some clinical trials evaluate supportive care for cancer patients, or assess methods of either preventing cancer or detecting it at an early stage, when treatment is often more effective. Supportive care studies that involve drugs, such as antiemetics, can be categorized by study phase like primary treatment protocols. Other supportive care studies, such as those studying psychological issues, are not phased. Prevention and early detection studies assess intervention and screening techniques; these studies may be directed at the population at large or towards people with known risk factors for cancer. STUDY DESIGNS (Back to top) Clinical trials are designed to answer questions concerning the safety and efficacy of treatment with statistical validity, particularly in phase III studies where the issue of replacing or augmenting accepted therapies or developing effective interventions to prevent cancer is at stake. Randomization is used to ensure an unbiased assignment to treatment. To more carefully balance the patient populations randomized to each arm, studies are usually stratified by factors known or expected to affect outcome. Such factors commonly include: patient age, histology and/or stage of disease, prior treatment, and presence or absence of prognostic factors. Sometimes phase II studies are stratified to assure a balance in patient entry despite limited accrual. Some clinical trials with multiple regimens have a crossover design to allow patients who are not responding to their originally assigned treatment to try the other treatment. Most phase II and III study designs include "early stopping rules" by which treatment is halted if too many patients experience unacceptable or unexpected toxicity or if treatment is significantly less (or more) effective than anticipated. INFORMED CONSENT (Back to top) Laws in the United States require that all subjects (or their guardians) on clinical trials sign an informed consent document indicating an understanding of the research in which they are participating. Protocols from other countries also must adhere to some type of informed consent, but laws vary and some countries permit oral instead of written informed consent. The informed consent should include an assessment of the risks and benefits of the study, the potential toxicities of treatments administered on study, other therapies that might benefit the patient, and financial responsibilities that will be incurred by participation in the study. Additionally, the confidentiality of patient records should be assured, and names and telephone numbers of appropriate people to contact with questions about the study are provided. Potential participants must also be assured that refusal to participate in the study will in no way compromise their access to other available treatment, and that if they choose to participate, they can leave the study at any time.
PHASE I TRIALS (Back to top) Phase I studies are the initial clinical tests of new treatments. The major purpose of a Phase I study is either to define a safe dose and schedule of agent or combination of agents or to evaluate the feasibility of combining treatment modalities. Phase I studies are usually designed based on promising preclinical data, such as in vitro cytotoxicity in tumor cell lines and safe administration with reproducible antitumor effect in animals, provided a stable and safe formulation of the agent is available.
Patient selection (Back to top) Generally, patients eligible for phase I studies have a confirmed malignancy for which no effective therapeutic options are known. Since the emphasis of the study is to determine a safe dose and not to follow patients for efficacy, many phase I studies allow entry of patients with any type of cancer. If an agent is thought to have activity in a particular cancer type (e.g., solid tumors, gastrointestinal tumors, or brain tumors), entry may be confined to patients with that type of cancer. Each protocol specifies criteria for patient eligibility. Typically, these criteria involve patient organ function and the extent of prior therapy. Good organ function is usually required since pharmacologic mechanisms of metabolism and excretion are often unknown in these clinical trials. Patients with seriously impaired kidney, liver, or lung function may have less tolerance to the study agent, making it difficult to assess toxicity and increasing the risk of unacceptable toxicity in that patient.
Informed consent (Back to top) Patients who enter phase I studies need to be fully informed that these studies usually represent the initial clinical experiments in humans. Expected side effects as noted in preclinical animal studies are presented, and the possibility that other, unpredictable side effects can occur must be stated.
Treatment approach and endpoints (Back to top) Most phase I studies treat cohorts of patients (typically, three to six) at predefined dose levels. The studies start at very low doses that were minimally toxic in animals (e.g., 1/10 x LD10 in mice). If that dose is safe, a cohort of patients is treated at a higher dose, with escalation continuing until a maximum tolerated dose (MTD) is defined. Various escalation schemas with successively smaller increases in dose are used, with rapid escalation at lower, presumably more tolerable, doses and slower escalation as the dose increases. The toxicity defining the MTD is based on toxicity found in animal testing, with provisions for unexpected toxicities. A dose for phase II studies is decided based on the MTD; usually, the phase II dose is about 80% of the MTD. Phase I studies may also be designed to determine the toxicity of lengthening or shortening an infusion schedule or the interval between treatments, assess supportive care designed to reduce toxicity, and determine an optimum biologic dose (OBD, i.e., the dose associated with the maximum desired biologic response for which toxicity is still acceptable). The schedule of drug administration in phase I studies is determined from preclinical testing. Common schedules include administration once every two to four weeks, administration for five consecutive days every three to four weeks, or daily dosing. Schedules may also be based on maintaining a certain drug level in the blood. Scheduling may involve varying the length of intravenous infusion. Commonly, a study employs one schedule; however, some studies compare multiple schedules within the same study. Agents may be given intravenously, orally, intraperitoneally, intravesicularly, or through other routes. Sometimes, investigators will randomize patients to multiple doses, schedules, or routes of administration to more quickly assess tolerance to various dosing routines and to rapidly follow up on the more promising one. Agents that are commonly used by one route or on an established schedule may return to phase I testing in order to determine a safe dose by another route or on a different schedule because an investigator has an important lead about how to improve the agent's effectiveness or reduce its toxicity. Also, phase I studies are employed to test the feasibility of combining drugs or different modalities of treatment (for example, chemotherapy with radiotherapy).
Required testing (Back to top) Because the purpose of phase I studies is to determine a tolerable dose and schedule, blood and other laboratory tests are performed at frequent intervals to monitor treatment effects on the function of major body systems. Additionally, blood levels of the agent are measured in order to establish the concentration and rate of disappearance of the agent from circulation and to correlate blood levels with side effects. This information helps researchers decide how to administer the agent in phase II studies and in patients with impaired organ function. For biological response modifiers, it is important to assess the effect on the immune function of the patient, particularly when the antitumor effect appears to be mediated through alteration or stimulation of the immune system.
Anticipated benefit (Back to top) While Phase I studies are primarily directed toward establishing a safe dose for further testing, there is always the possibility of therapeutic benefit in these studies.
PHASE II TRIALS (Back to top) Phase II studies assess the antitumor efficacy of a new cancer agent, a new combination of agents, or a new modality of therapy. Using the dosage (dose and schedule) found to be safe in phase I studies, the new treatment is given to groups of patients with one type of cancer or related cancers. Occasionally, agents with preclinical efficacy against a broad range of tumors are tested in phase II studies that accept patients with a broad range of cancers. Further determination of toxicity is the second major purpose of phase II studies. In most cases, fewer than 100 patients have ever received the new treatment prior to phase II testing. Phase II studies normally involve several trials that collectively treat hundreds of patients with many types of cancer. Unusual or chronic toxicities, which are often missed entirely in phase I testing, may appear during phase II testing; this is considered when deciding whether the agent should be further evaluated in phase III studies.
Patient selection (Back to top) Patients enrolled on phase II studies are usually required to have measurable biopsy-proven disease, to have a good performance status, and to be free of serious intercurrent disease. Nearly normal renal, hepatic, cardiac, bone marrow, and pulmonary function are usually required, with exceptions sometimes made if an organ system is involved with tumor. Often, patients treated with phase II agents must have failed treatment with other chemotherapy, surgery, or radiotherapy. A number of studies have demonstrated that patients who have received extensive prior therapy are unlikely to respond to phase II agents that are effective in minimally pretreated patients. Therefore, many phase II studies restrict the amount of prior therapy to one or two prior regimens. For many disseminated or recurrent solid tumors, no effective therapy exists and a phase II study should be considered as first-line treatment.
Informed consent (Back to top) Patients entering clinical trials must be counseled in detail about their prognosis and chances for being treated effectively with other investigational or standard therapies. The known side effects of new cancer agents must be conveyed to them. In addition, they must understand that unexpected side effects may occur. While tumor response, increased survival, and cure are all potential benefits of phase II therapy, investigators must convey a realistic assessment of the likelihood of any of these benefits occurring.
Treatment approach and endpoints (Back to top) Usually between 25 and 50 patients are treated for each tumor type and for each dose and schedule selected from phase I testing. The decision on which cancers to treat may be based on preclinical efficacy against implantable tumors in animal models and in human tumor cell cloning assays or from hints of activity in phase I studies. In other cases, patients with tumors responsive to similar analogs will be treated with the newer drug. Other phase II studies accrue patients with common adult solid tumors who are not eligible for other treatments of known efficacy. Most phase II studies require patients to have measurable or evaluable primary tumors or metastases. Measurable disease is serially measurable in two dimensions and includes the following: nodules on chest x-ray that are confined to the parenchyma, freely movable lymph nodes or masses, masses imaged on CT or MRI, and skin nodules. Examples of lesions that do not qualify as measurable are diffuse bony metastases on bone scan, pleural effusions, patchy pulmonary infiltrates, poorly defined subcutaneous masses, and ill-defined abdominal masses. Measurable disease is necessary to objectively define tumor response. A complete response means the disappearance of all known sites of disease without the development of any new disease for at least one month. Partial response is at least a 50% decrease in the sum of the products of the bidimensional measurement of all lesions with no new disease appearing for at least one month. Patients with ill-defined sites of disease cannot be accurately evaluated according to these criteria. Some studies will also allow enrollment of patients with evaluable disease. Evaluable disease may be measurable in one but not in two dimensions. Some studies consider any indicator of disease that can be followed serially as suitable for evaluability, including serum tumor markers and ascites. The dose and schedule for phase II studies are taken from phase I studies. In the absence of unacceptable toxicity, some phase II protocols allow patients to continue treatment as long as their tumors respond or remain stable; other phase II protocols limit therapy to a specified number of treatments. Following treatment, patients are usually followed for outcome, i.e., for disease recurrence or progression, for survival, and for evidence of late toxicity. An adequate number of patients needs to be treated to determine the level of drug activity. If no responses are seen in two trials, the agent is usually considered inactive for that tumor type. However, if the patients in these studies have received extensive prior therapy or failed to receive adequate doses of the new agent, negative results are less convincing. If responses are seen, the number of patients studied is increased to more precisely define the response rate. Some clinical trials combine phase I and phase II study designs within the same protocol, and patients treated at the recommended phase II dose during the phase I portion may be added to the statistical pool for the phase II portion. Newer statistical approaches to phase II studies include randomization of patients to two or more treatment arms, each testing a new agent. The agent producing the highest response rate, even if the response rate is not significantly higher than that produced by other agents, is chosen for phase III testing.
Required testing (Back to top) Since patients entered on phase II studies may be receiving a new agent, careful observation is mandatory and requires serial laboratory evaluations to monitor for organ damage. Tumor measurement requires a detailed history and physical examination, blood tests, x-rays, scans, and sometimes repeated surgical biopsies. Patients must be informed of the nature of the diagnostic monitoring required by the study as part of their informed consent.
Anticipated benefit (Back to top) The antitumor efficacy in humans of any new phase II drug is unknown at the outset, but benefit for some patients is hoped for based on preclinical data. Phase II agents that demonstrate partial response rates of at least 15% may be considered for further testing, often combined with other agents of proven efficacy. New combinations are pilot tested in other phase II studies and may include only standard agents or both standard and new agents. A number of potentially effective agents are discarded after serious toxicity is noted in phase II studies. Ethical considerations require that investigators be willing to terminate phase II studies when severe toxicity without compelling efficacy is observed. Once a new agent has demonstrated adequate efficacy and safety in phase II studies, it enters phase III testing where it can be compared to generally accepted therapies for efficacy and toxicity.
PHASE III TRIALS (Back to top) Phase III studies are designed to compare one or more treatments. A new drug or drug combination may be tested against one of proven efficacy. Phase III studies often have multiple endpoints. Overall and disease-free survival are nearly always endpoints; differences in response rates, toxicity, patterns of recurrence, and quality of life might also be endpoints. At the conclusion of a properly designed and conducted phase III study, the new drug or drug combination will be found to be inferior, equivalent, or superior to the standard treatment with respect to the major endpoints. The degree of difference will be known and statistical significance will be estimated.
Patient selection (Back to top) Generally, patients eligible for Phase III trials must have good performance status, be free of intercurrent disease, and have normal renal, hepatic, cardiac, bone marrow, and pulmonary function. Phase III studies may exclude patients who have received prior therapy for their cancer, or they may enroll only patients who have had prior therapy. Studies that are designed to assess response rate require disease that is bidimensionally measurable or at least evaluable (capable of being objectively assessed for improvement or worsening over time).
Informed consent (Back to top) Patients entering phase III studies must be counseled in detail about their prognosis and the expected benefits from participation in the study, as well as benefits that could be expected from treatment with other investigational or standard therapies. The known side effects of anticancer agents must be conveyed, and they must understand that, despite prior phase I and II testing, unexpected side effects may occur. The expected length of the study and follow-up measures should also be made clear.
Treatment approach and endpoints (Back to top) Patients are randomly allocated to the treatment options to help ensure unbiased comparison of the treatments. The numbers of patients who are treated with each regimen depends on the number of major events (usually recurrence or survival) predicted to occur during the course of accrual and follow-up. Anywhere from 100 to 1,000 or more patients may be required, and accrual may take 2-5 years or more. Studies must be designed to account for major prognostic categories by either entering large numbers of patients or by stratification on the basis of known prognostic factors. For example, a phase III study in breast cancer may randomize premenopausal and postmenopausal women separately. In spite of stratification and randomization, some studies will enroll more patients with good prognosis in one arm than in the others. In such cases, statistical techniques exist to retrospectively "balance" treatment arms for prognostic factors. The dose and schedule for the new arms are based on phase II studies. Dose and schedule for the standard arm are taken from published literature or prior phase III studies. In a few phase III studies, historical data are used instead of a standard arm and all patients are allocated only to investigational arms. There is considerable controversy about the validity of using historical data versus concurrently randomized controls in phase III studies. After completion or discontinuation of treatment, follow-up of patients typically lasts for life or until other treatment is initiated. The requirement for large enrollment and lengthy follow-up makes phase III studies expensive and difficult to run. Therefore, many phase III clinical cancer trials are run by several institutions under the organizational structure of cooperative oncology groups funded by the National Cancer Institute or by drug companies.
Required testing (Back to top) Close observation of patients for both acute and chronic toxicity is necessary; unusual or chronic toxicities may be observed only in the large number of patients on phase III new regimen arms. Tumor measurement requires a detailed history and physical examination, blood tests, x-rays, scans, and sometimes repeated surgical biopsies. Patients must be informed of the nature of the diagnostic monitoring required by the study prior to their giving informed consent.
Anticipated benefit (Back to top) Phase III studies offer the most up-to-date treatment for a given indication, whether the studies are for primary cancer treatment or supportive care.
|