Initial human trials with an investigational new drug (phase I and 2) : planning an management.
The planning and management of the initial human trials with each investigational new drug (I.N.D.) presents a fresh and significant challenge to the skills and competence of a clinical pharmacologist and also to the expertise of his interdisciplinary colleagues. Unlike phase 3 trials,, , , , ,  there is relatively scarce published literature on the guidelines for the conduct of phase 1 and 2 studies., , ,  As a consequence, there are sometimes inordinate delays in the proper planning and execution of phase 1 and 2 clinical trials. Sometimes certain defects in the initial human trials are censored either by the drug administration or by other experts; additional studies are demanded and further delays in the development of a new drug become inevitable. It is desirable at the outset to adopt, even for phase 1 and 2, the attitude suggested for the controlled trials by Sir Austin Bradford Hill, "In general, it will be seen that the essence of a successful controlled clinical trial lies in its minutia in a painstaking, and sometimes very dull, attention to every detail".
The early limited human trials with a new investigational drug has been designated as clinical pharmacology in paragraph (a) (2) of section 130.3 of the regulations of the Food and Drug Administration of the United States of America. This paragraph is as follows: "Clinical Pharmacology. This is ordinarily divided into two phases:
Phase 1 starts when a new drug is first introduced into man-only animal and in vitro data are available with the purpose of determining human toxicity, metabolism, absorption, elimination and other pharmacological actions, preferred route of administration, and safe dosage range; phase 2 covers the initial trials on a limited number of patients for specific disease control or prophylaxis purpose. The first two phases may overlap, and when indicated, may require additional animal data before these phases can be completed or phase 3 can be undertaken".
Phase 1 studies can be carried out either in volunteers or for certain diseases in patients with minimum disease who are likely to be benefitted by the new drug. The studies are carried out usually in a well-equipped Clinical Pharmacology Unit of a teaching hospital or a research institute with the facilities for common resuscitative measures. It is desirable that expert supervision of medical specialists, clinical pharmacologists and trained house-staff, technicians and nurses should be available during phase 1 studies. The objectives for phase 1 studies are: (1) human tolerability of the compound (2) pharmacokinetic behaviour and metabolism of the new investigational drug (3) effects of the compound on physiological functions (4) preferred route of administration and (5) safe dosage range. For the phase 2 studies, in patients, the additional objectives will be: the assessment of therapeutic or prophylactic activity and efficacy, optimal dosage range, an evaluation of the drug-emergent side-effects and toxicity, the choice of the best pharmaceutical formulation and probably studies to determine the mechanism of action.
The detailed pre-clinical data on a new compound should be available for a well informed and intelligent planning, design and conduct of early trials. It is also essential for the ethical and scientific justification 20 of human exposure to a new compound. Clinical pharmacologists embarking on the phase 1 and 2 trials of a new compound participate in the first planning workshop where all the experts who conducted the pre-clinical studies meet and discuss the relevant points in pharmacology, toxicology and metabolism of the new investigational drug. Such a workshop would discuss the design and nature of experiments to be carried out during phase 1 and 2 trials; protocols and case record forms can also be evolved.
The pre-clinical data are collated and prepared for presentation to the drug regulatory agencies and to the clinical pharmacologists. The perspectives of the regulatory authorities and the clinical pharmacologists, though emphasized to be different, do overlap to a considerable extent. The following pre-clinical data on an investigational compound would serve both as a documentation for I.N.D. application to the Drugs Controller as well as a basis for the clinical orientation for the participating clinical pharmacologists.
1. Chemistry and physico-chemical properties
This should cover the chemical structure, chemical and generic name or a serial number, solubility data, physical form-crystalline or amorphous, colour, taste, etc.
2. Formulation data
This would describe the type of formulation e.g. tablet, the strength of active ingredients and excipients, the particle size, the disintegration and dissolution rates, the content variability, stability data and the stipulated conditions for storage facilities.
3. Details of the method of manufacture and analytical specifications
These data are mainly required by the drug-regulatory authorities; a clinical investigator may ask certain details in case he suspects that a contaminant residue during manufacture may present a clinical hazard.
4. General pharmacodynamic profile
This should cover the pharmacological actions of the new compound on isolated tissues, organs or in intact animals.,  The effects on the cardiovascular, central nervous and the gastro-intestinal systems are usually carefully studied. Interactions with drugs likely to be used concomitantly are also studied in animals. Depending on the nature of the compound special experiments are designed. Missing but vital data can be requested by a clinical pharmacologist and may have to be generated even prior to phase 1 studies. The cumulative frequency of such requests reflects the degree of vigilance exercised during the scrutiny of pre-clinical data.
5. Specific pharmacological efficacy
Full reports on the in vitro or in vivo data demonstrating the efficacy of the new compound for a specific therapeutic or prophylactic indication have to be presented. It is also desirable that data demonstrating the advantages over the existing standard drugs are presented. The target action, E.D.50, activity in specific animal disease model, etc. must be exhaustive. The mechanism of action should preferably be elucidated.
6. Pharmacokinetics and drug metabolism
The data in two species of animals on the absorption, distribution, metabolism and excretion of the radio-isotope-labelled or the cold compound are now being generated even prior to phase 1 studies. The half-life values of the new compound, pathways of metabolism and the routes of excretion assist the rational design of phase I studies viz. safe dosage range and pharmacokinetics. The nature of metabolites-active, inactive or toxic also are determined in one or two species and are of value for the clinical activity and safety of the compound.
7. Toxicological studies
The objectives of pre-clinical toxicity studies are (1) to assess the safety of the compound by knowing the maximum tolerated doses in several species-rodent and non-rodent (2) to know the toxic potential of the compound and the nature of adverse reactions by pushing the doses to the limit in two species and (3) to determine the therapeutic safety of the compound by comparing ED50 to LD50 or ED50 to M.T.D.50 (Minimum Toxic Dose).
These goals are fulfilled by acute, subacute and chronic toxicity studies. Usually, prior to phase 1 and 2 studies, only acute and subacute toxicity data are available. [Table 1] shows the animal toxicity requirements for the different phases of clinical trials according to the categories of the routes of administration and the duration of treatment. Teratogenic, mutagenic and carcinogenic studies are not usually carried out prior to phase 3 studies. A clinical pharmacologist must carefully study all the details of the toxicity studies. The problems of extrapolation of the animal data to man are complex and have been extensively discussed,  However, it is interesting to note that animal data are not so redundant for human safety as often mentioned in the reviews. The toxic effects observed in man and which cannot be readily elicited in animals are drug allergy, drug idiosyncrasy and pharmacogenetic drug reactions. But during phase 1 and 2 trials one is unlikely to detect these rare reactions.
The legal aspects,  involve filing an application to the drug regulatory authorities with the following: (1) All the documents of the pre-clinical data (vide supra) ; (2) The plans, protocols and case record forms for the phase 1 and 2 studies; (3) Name, address and biodata of each clinical investigator; (4) An agreement from the sponsor that the drug regulatory authorities will be notified of adverse reactions which arise during either the ongoing animal or human studies; (5) The nature of "informed consent" and (6) An agreement to submit annual progress reports. A copy of the letter from the drug regulatory authority granting permission for phase 1 and 2 trials should be on the files of the clinical pharmacologist. It is also desirable that an insurance coverage is made available both for the investigator as well as the volunteer/patient. Unfortunately, in India, the general insurance companies are not aware of this need and are generally unwilling to create fresh new policies for the liability arising out of adverse drug reactions.
The ethical aspects of clinical drug research in man have engaged the minds of specialists from diverse fields-medicine, law, religion, etc.., The project and protocols must be approved by the ethical committee of the institution where the studies are planned. Informed consent must be voluntary and a consent form must be prepared for the individual project and signed by the volunteer/patient, the investigator as well as witness.,  The form should contain the purpose and a brief description of the project; the procedures involved, expected discomfort and/or risks involved. There must also be a statement that the volunteer/patient will report any untoward effect at the earliest and that he or she has a right to withdraw from the trial at any stage. The agreement part must cover free choice and a statement of having understood the nature of the experiment. The compensation to be paid to the volunteer/patient for the trial has to be commensurate with the discomfort involved, number of venipunctures and the loss of daily wages, transport money, etc. Individual institution should evolve guidelines for the payment to the volunteers. Neither excessive payment nor the complete lack of payment should be encouraged because the type of personalities drawn to these two ends of the spectrum may not be suitable for early drug studies.
The nature of design of the phase 1 and 2 studies is dependent on the objectives to be fulfilled within a specified range of time and with the resources available. The protocols and the case record forms (C.R.F.) have to be carefully and specifically evolved for each individual project fulfilling one or several objectives. The specific objectives for every project must be precisely worded. As Uberla has put it, "Part of the difficulties and the frustration of phase 1 studies originates from the fact that the goals are conflicting, not sufficiently specified, and the relative weights are not set beforehand." It is desirable that extensive literature review is carried out on the related drugs prior to the design and protocol development for the new compound.
The first studies with an I.N.D. are usually carried out in normal male healthy volunteers. At this early stage, almost each volunteer study serves as a detailed experiment. The protocol and design should be fixed in its care-structure of imperative clinical and laboratory monitoring but flexible enough to pursue the novel observations arising during the study. Placebo may have to be included to provide control observations; it may be also essential to re-challenge for the assessment of early side-effects. The experimental design must be kept simple involving usually the volunteer/patient studies, depending on the type of the compound. The protocol for phase 1 and 2 studies for each specific project should cover the following points:
1. Title of the project.
2. Investigators-Names, qualifications, designations, addresses, telephone numbers and specific responsibilities in the project.
3. Institution-specific location in the institution (preferably a map with the floor diagram).
4. Objectives of the study-Precisely worded goals.
5. Criteria of selection and exclusion of volunteers/subjects and the sample size.
6. Drug-therapy-investigational new drug, placebo and other drugs, dosage forms, routes of administration, frequency, quantity of dosage, concomitant drugs to be avoided, etc.
7. Criteria and methods of evaluation of efficacy and safety.
8. Design of the study with flowcharts Nature of controls, randomization, execution of the studies, methods of analysis.
9. Informed consent form.
10. Case record form-physical characteristics, duplication, a model filled form, space for additional- notes, mailing or collection of forms, etc.
11. Expected duration of the trial.
12. Signature of the investigator and date.
Selye classified the following categories of qualifications of the scientist: (1) Enthusiasm and perseverence; (2) Originality-creativity, the spirit of invention, imagination, intuition, genius; (3) Intelligence logic, memory, experience, concentration, abstraction; (4) Ethics-honesty with self; (5) Contact with nature-observation, technical skill; (6) Contact with people-understanding of self and others, compatibility, organization of teams, convincing others and listening to argument. It would be almost impossible to find an investigator who would fulfill all the above pre-requisites. However, many of the aforesaid qualifications are essential for phase 1 and 2 studies with a new compound.
Imhof has discussed, in detail, the problem of the choice of the investigator. In general, phase 1 and 2 should be conducted by an experienced, well-trained and motivated clinical pharmacologist; phase a studies are usually entrusted to clinical investigators-experienced specialists who conduct such studies in close collaboration with the clinical pharmacologist, who conducted phase 1 and 2 studies. In India and other developing countries, there is an acute shortage of experts, who can conduct phase 1 and 2 studies. As a consequence, only a few investigators get overburdened and the quality of phase 1 and 2 studies may suffer. It is desirable that even an expert clinical pharmacologist with excellent ward and other facilities does not take up more than 3 compounds concomitantly for phase 1 and 2 studies.
It is a common practice to include healthy adult male volunteers for phase 1 studies. The medical students, members of the research staff and prisoners have often been volunteers for such studies. , ,  However, it is not easy to hospitalise healthy adult volunteers for one or two weeks. In India, several phase 1 studies have been conducted in hospitalized patients, who are convalescent from minor intercurrent illness and/ or in professional blood donors. Even moderately ill patients, who are likely to benefit from the drug, can be utilized for phase 1 studies.
Children, women of child-bearing age and the old persons are not generally included in phase 1 studies, unless the nature of the investigational new drug necessitates their inclusion. For example, in the early evaluation of an oral contraceptive, women of child-bearing age, who have undergone tubal ligation are often selected as volunteers, because of obvious need to study the effect of the new agent on the menstrual cycle, ovulation, hormonal profile, etc.
The criteria of selection of volunteers and/or patients must be carefully laid down in the protocol and must be strictly adhered to. The age-range, sex, bodyweight, height, mental development, and the screening criteria for health must be specifically mentioned for the healthy volunteers. For the patients, the grade of severity of the disease, the criteria for diagnosis, absence of complications and absence of any severe renal, hepatic, cardiopulmonary or nervous disease or allergy must be specified. The later precautions are very essential for phase 2 studies too.
The number of subjects must not be left totally to the judgement of the investigator. A statistician must be consulted for the adequate sample size for each project. For phase 1 studies usually 10 to 20 volunteers suffice for safety, activity and pharmacokinetic profile of I.N.D. For phase 2 studies, the number of patients varies as per the design of the trial. Dose-searching (phase 2a) usually requires 20 to 30 patients. Further trials (phase 2b) may require additional 30 to 50 patients. This is a general guideline and the nature of the investigational compound may influence the sample size.
One of the most controversial areas in clinical pharmacology is the choice of the initial human dose. The recommendations for calculating an initial dose in man have been numerous and widely divergent; the suggestions usually involve some fraction of LD50, maximum tolerated dose, or minimum toxic dose-calculated on the basis of surface area or body weight., 
Dollery and Davies suggested that 1-2% of the dose predicted from animal data can be used. The increments in single dose in the subsequent volunteers were proposed to be very small with a limitless supply of volunteers but can be larger with a limited number of volunteers. This is a bit confusing. Severinghaus suggested that the first human dose may be planned upon the basis of the minimal effective dose per kilogram in the dog; use of the dog dose or less as a starting point increases safety. This also sounds vague.
Pitts, after a careful analysis of the predictor curves for phase 1 studies based on maximum tolerated dose by weight or surface area, has evolved certain recommendations:
(a) Theoretical maximum dose in man calculated as one-fifth or one-tenth of the maximum tolerated dose in the most sensitive species and assuming an average 70, kilo man.
(b) An estimate of probable daily human therapeutic dose from the animal pharmacology or by reference to related standard agents.
(c) Prior human experience with agents in a similar or related chemical class.
(d) Utilising the formula Y=A+B-vx (based on surface area) and applying the 25 per cent probability factor, ED25 is calculated.
Balancing these factors one against the other, a starting dose can be selected low enough to avoid unnecessary risk. In general, the first dose has to be related to the ED50 in the animal model of the disease. The first dose can be 1/10th of the projected therapeutic dose in man. However, this can be even 1/5th if the toxicity in several species suggests very good safety of the compound. For example, with a new benzimidazole with a projected therapeutic dose of 2 mg/kg in man, the initial human dose can be 0.2 or 0.4 mg/kg. A high maximum tolerated dose permits the choice of a higher fraction of the estimated therapeutic dose in man. There is an element of intuition in the choice of the first dose. This is based on the extensive experience of the clinical pharmacologist. It may be desirable to consult senior experienced colleagues in clinical pharmacology for specific problems.
The increments in dose in the first 3-4 volunteers have to be very cautious and based on the on-going experience in tolerability. It is preferable to study single doses in the first group of volunteers. The increase in dose can be 1.5 or 2 times the previous dose depending on the pharmacodynamic effects of the previous dose and the margin between the projected therapeutic dose and the theoretical maximum human tolerated dose. Availability of pharmacokinetic data in the first group of volunteers, assists the rational increments in single and multiple dosage, based on the CMAX, TMAX and half-life of the compound.
The toxicity data of the new compound in animals and the tolerability profile of the chemically and biologically related drugs assist in planning the safety studies in man. The drug and past disease history of the volunteers should be carefully taken; allergic reactions, side-effects of drugs, sub-clinical or chronic illnesses may often be revealed by a meticulous history, physical examination and investigations. The need to obtain baseline symptoms cannot be over-emphasized. The check-list should include all the known side-effects of related drugs. Baseline record would distinguish the `non-drug adverse reactions'.
It may be desirable to obtain 3, 2, 1, pre-drug days' data on vital signs such as blood pressure, temperature, pulse rate, respiration rate, etc. three times in a day prior to drug administration so that basal fluctuations are recorded. Immediately before and after drug administration, blood pressure, pulse rate, E.C.G., etc. are studied at frequent intervals (½ to 1 hour). Treatment-emergent symptoms and/or signs are carefully recorded with all the details of onset, duration, severity, measures undertaken, etc. Special studies may have to be planned for specific expected side-effect of a drug e.g. salivery volume with an anti-depressant. It is desirable to review the literature on wide-effects and adverse reactions of drugs related to the compound; this assists well-planned and meticulous clinical tolerability studies.
The basal and post-drug monitoring of organ functions by appropriate laboratory tests, E.C.G., etc. ensures an early detection of often sub-clinical drug-induced changes. [Table 2] lists several organ functions monitored during phase 1 and 2 studies. All the tests are not mandatory; the nature of compound may influence the choice of tests. The laboratory investigations are commonly carried out basally, during the drug administration and at the end of therapy. Residual serum or plasma may be frozen for a repeat check with different methods. Abnormal values may demand more frequent follow-up tests till normal values are obtained.
The early studies with a completely new compound, with structural novelty, may have unpredictable effects in man.; these, for species-differences, may not have been observed in animal experiments. It is essential to be in readiness to meet the challenge of any unforeseen hazard or adverse reaction. [Table 3] lists the essential drugs and other equipment, which must always be on hand to treat an emergency. The investigator must also have specific antidotal measures for the compound under study. The resuscitation equipment must be kept functional by frequent service and mock-run; the drugs and injections should be periodically checked for the date of expiry.
The tolerability studies in volunteers include single and multiple rising dose studies both in and above the therapeutic range. It is desirable to combine pharmacokinetics with tolerability studies. This may provide data about toxic plasma levels of drugs. Rechallenge doses for side-effects are usually not favoured during phase 1 but may occasionally be possible in phase 2 studies. However, placebo rechallenge is desirable to exclude suspected placebo adverse reactions. During phase 2 tolerability and efficacy studies in patients, placebo usage is advisable.
Any moderate or severe reaction observed during phase 1 and 2 studies must be investigated thoroughly. The investigator must inform the drug regulatory authority and the sponsoring pharmaceutical company. Additional animal pharmacological, biochemical or toxicity studies may have to be carefully designed from the clinical feedback of the adverse reactions. For idiosyncratic reactions, volunteers have to be screened for certain genetic conditions, e.g. G-6-P.D. deficiency has to be ruled out in case an analogue of primaquine is being subjected to early clinical trials. The new compound must be studied from the point of view of chemical kinship to other drugs known for pharmacogenetic reactions.
The design for the pharmacodynamic experiments depends on the biological profile of the IND and the criteria chosen to monitor the actions on physiological functions. Certain pharmacodynamic effects are easily measurable by the changes induced in the pulse rate, respiration, blood pressure, body weight, etc. Usually all phase 1 trials cover these parameters. However, dose-response curves can be evolved by a careful monitoring of well-defined clinical, biochemical and electrophysiological end-points.
Human pharmacology laboratory (HPL) has to be an integral part of a clinical pharmacology unit in a teaching hospital or in a research centre. At such HPL, a battery of non-invasive tests can be evolved to study pharmacodynamic Effects of new compounds in volunteers. It is desirable that a clinical pharmacologist acquaints himself with non-invasive techniques used for various physiological functions, e.g. systolic time intervals. The latter can be studied by a simultaneous recording of E.C.G., phonocardiogram and carotid pulse wave on a three channel polygraph,  Cardioactive drugs, B-adrenergic blockers and even other drugs have been studied meaningfully by such techniques., ,  Heart rate as a parameter of measurement in phase 1 studies provides interesting data on the activity of the new compound. The influence of B-adrenergic blockers and other cardioactive drugs on the heart rate and rate-pressure product has been studied by several investigators.,  Such studies with drugs not having a direct cardiovascular action may also be carried out during phase 1 studies.
Diverse methods have been employed in human pharmacology:
(1) Erythema induced by UV light and its delay by anti-inflammatory agents.
(2) Effect of vasodilator drugs on the fore-arm arteries and dorsal hand veins.
(3) Effects of drugs on buccal mucosal potential difference.
(4) Changes in nasal mucosal pH with decongestants.
(5) Alterations in human skeletal muscle lactate.
(6) Techniques of studying cerebral blood flow in man.
(7) Techniques to measure coronary blood flow in man.6
(8) Pupillography and pupillometry to study drug interactions.
(9) Drug studies with sustained handgrip.
(10) Galvanic skin resistance measurement.
(11) Nerve conduction velocity determinations, etc.
The list can be even much more elaborate. It is given here to emphasize that innumerable techniques are available to design specific pharmacodynamic experiments appropriate for the new compound. A sophistication in human pharmacological methods will permit an early and objective evaluation of pharmacodynamic effects of a new compound in man.
The objective end points-serum and; or plasma levels of substrates, metabolites, electrolytes, hormones-may also be studied as criteria for pharmacodynamic effects. The hormonal profile during studies with a new drugs can provide interesting insights about a potential antifertility agent. The analytical methods used for pharmacodynamic experiments have to be highly specific, quite sensitive and requiring very small blood samples. The new field of human bio-chemical pharmacology offers many opportunities for interesting work with new compounds.
The pharmacokinetic and metabolic studies,  with an investigational new drug are strongly recommended during the first human trials, with dose-ranges permitting the measurement of the compound and its metabolites. The knowledge about the plasma half-life in man, the nature and magnitude of the plasma level curve, completeness of excretion, etc. would permit rational dosage schedules to be evolved.. The studies, as pointed out earlier, can be correlated with the pharmacodynamic or toxic effects.8,  The design of the protocols and case record forms for pharmacokinetic studies needs a close collaboration between an expert in drug metabolism and a clinical pharmacologist. Ritschel has well covered the diverse aspects of planning, executing and evaluating the pharmacokinetics during the early studies in man. Pilot studies with intravenous or oral single dose of the new compound are carried out with multiple blood samples, quantitative urine and stool collections. Subsequently, wish multiple doses, steady state levels of the compound are measured on several days during therapy and continuously for 3 to 5 days after the end of therapy. Both volunteers as well as patients can be employed for these studies. Special studies like antipyrine half life, measurement or typing the acetylator status may be carried out if indicated., 
Radioactively labelled drug studies are being carried out with decreased frequency since analytical methods employing gas chromatography, high performance liquid chromatography, stable isotopes, etc. have provided very sensitive and specific measurements., , 
The phase 1 experience in volunteers provides guidelines for rational dose searching in a limited number of patients. It is desirable to have clear objectives for dose-searching studies. A new drug has to offer several advantages over the existing drugs. The following points are to be considered prior to embarking on dose-searching: (1) The potency of the new compound-effects at smaller doses; (2) A reduced frequency of the drug administration-single or two doses per day; (3) A shortened duration of therapy -one or two days instead of 5-7 days with a standard drug; (4) Wide therapeutic margin-optimal efficacy with least possible side-effects; (5) Elimination of constraints of therapy-with or without food or relationship to beverages etc. (6) More acceptable routes of administration-oral route preferred and (i) Improved palatability or presentability of the new compound-chewable tasty formulations.
It is desirable to attempt to arrive at the smallest effective dose, given at the largest possible interval for the shortest duration. The schedules of other related drugs must be carefully studied for these experiments. The end-points of efficacy must be well-defined. Objective methods of assessment based on clinical, biochemical or electro-physiological criteria should be preferred to vague rating scales during the phase 2 studies. However, for the subjective responses, it may be desirable to include the validated rating scales.
On each dosage schedule (quantity, frequency and duration), about 6 patients may provide sufficient data. If 4 out of 6 patients show excellent efficacy, the schedule may be continued further. But if this schedule is found ineffective, retaining the frequency of administration and duration of therapy, the individual dose of the drug, etc. may be increased by a small magnitude-by one-fifth or one-fourth. The increments can be based also on pharmacokinetic data. When excellent efficacy is established, subsequent attempts may be made to optimize the frequency of administration and the duration of therapy. Too much enthusiasm for reducing the duration of treatment must be curbed for diseases which require hospitalization and careful observation. For example, the current trend to cure amoebic liver abscess by a single day or two-day course may not be justified in view of the serious nature of the disease and a need for supervision. It may be worthwhile to combine dose searching studies with pharmacokinetic and tolerability studies. This would provide the range of therapeutically effective plasma levels of the compound, as well as correlation with toxic drug levels, if any. Usually, steady state levels of the drug provide useful data. Metabolism of the drug, may also guide the dose-searching experiments. For example, occurrence of an active metabolite may reduce the frequency of administration and the half-life of the active metabolite has to be considered while altering the dosage schedule. A drug, predominantly metabolized by acetylation, may require that acetylator-phenotyping be carried out in patients prior to the dose-searching studies.
Sometimes dose-searching may have to be continued even during phase 3 trials. This is particularly so, when two dosage schedules appear equally effective and only a large number of patients may provide data for the preference of one over the other. Usually dose-searching studies require thirty to fifty patients; this number is small and as patients with severe disease are not often included during phase 2, it is desirable that such patients too are studied during the phase 3 trials.
Statistics for phase 1 and 2 trials
There are three aspects of statistics which have to be considered for phase 1 and 2 studies; (1) the nature of the experimental design, with the type of controls, (2) elimination of investigator's bias and (3) the methods of statistical analysis for knowing the significance of the data.
During phase 1 studies, design for a single subject poses problem. However, it has been proposed that sufficient number of repeat measurements before and after drug may provide data for analysis of variance. It is desirable that the number of measurements are equal before and after treatment. If the F-test is significant, the difference between the measurements before and after treatment is considered larger than the variation between the replications. Whether the difference, at the specified level of significance, is ascribed to the drug is a matter of clinical judgment and experience. Latin square design can be employed with a number of subjects, doses and days of therapy being equal. The design of early phase 2, trials also follows similar patterns of within-patient comparisons. At a later stage in phase 2, after dose-searching, standard controlled trial designs can be incorporated There is a need for research in novel experimental designs for phase 1 and 2 studies.
For the elimination of the investigator's bias, randomization chart prepared from "Tables of Random Sampling" are used during late phase 2 studies. However, for phase 1 and early phase 2 studies, it is desirable to have two experienced investigators agreeing for the choice of the subject, based on well-laid down criteria. Sometimes stratification may have to be used as to age, sex, duration or severity of disease, etc. if these factors are expected to influence either the efficacy or the tolerability of a new compound. Placebo or dummy medication can also be effectively used to eliminate the investigator's and patient's bias. With objective criteria of response, the bias in clinical evaluation is minimized. It may be desirable to have two clinicians independently assess the patient and enter their scores separately.
For the analysis of the significance of the trial data, there has been scarce attention paid to phase 1 and 2 studies, as compared to phase 3 studies. The method of analysis is dependent on the nature of the experimental design, the type of data, and the mode of obtaining and calculating the parameters. The counted data may be analysed by the Chi-square test for difference between the percentages. With the small numbers during phase 1 and 2 studies, the ready-made tables can be gainfully employed. Non-parametric tests like Wilcoxon can also be useful. For late phase 2 trials, Student's t-test, paired t-test, analysis of variance are commonly used. These vests do not play the same role in phase 1 and early phase 2 studies One must consider each volunteer or patient during phase 1 and early phase 2 studies as a complete detailed experiment by itself, observing and noting down every unusual effect and being always on a look-out for the unexpected. Hypotheses generation is facilitated by in depth attention to the maximum collected data of a qualitative nature.
Accurate record of each experiment during phase 1 and 2 studies is mandatory not only for the scientific purpose but also for the statutory and other review requirements. The mode and time of entry of the specified data about procedures and findings must be discussed in detail. Once a standardized mode is agreed upon, it must be strictly adhered to. This would avoid any ambiguity in the future interpretation of data. The data should be available in duplicate or triplicate and filed at two or three different places to avoid the hazards of accidental loss. Whenever Xeroxing or microfilming facilities are available, it is useful to store the original case record forms in Xerox copies or micro-films. The volunteer and patient records must be serially filed with an appropriate index. This ensures ready retrieval for analysis or future reference. All the data need to be stored for many years beyond the life-time of the drug. Hence it is desirable to give due thought to storage space and conditions. There have been attempt's also to store the data on IBM punching cards for easy retrieval and analysis.
Phase 1 and 2 studies comprise of a series of experiments aimed at developing separately or simultaneously the pharmacodynamic, pharmacokinetic, tolerability and therapeutic activity profile of a new compound. The design, execution and analysis of such studies demand expert and innovative team work for several years from basic and clinical pharmacologists, specialists in medicine, biostaticians and biochemists, etc. At present, there is much art involved in such studies; research in methodology of phase 1 and 2 studies is urgently called for. However, the current shortage of clinical pharmacologists does not permit the time and other inputs for such research. Such efforts must be actively encouraged by concerted endeavor of all the sectors involved in new drug development. The development of reliable, efficient, speedy and sophisticated methods for phase 1 and 2 studies will shorten the clinical phase of new drug development, spare valuable resources and lead to rapid and reliable pre-market clinical drug studies.