The purpose of Phase I clinical trials is to evaluate safety, determine a safe dosage range and identify side effects of a new molecular entity. As a result, it is dosed above and beyond the estimated therapeutic dose. This means the formulation requirements are very different to a commercial-like formulation development where marketing-/patient-driven factors are important considerations.

For Phase I, a fit for purpose formulation is required. This is, one that will allow dosing across the relatively wide dose range, but without unnecessary time and cost being expended on optimization. Attrition in early phase clinical trials is well documented with the success of new molecular entities progressing from Phase I and Phase II around 65% and 32% respectively, according to Hay et al.¹ For this reason, an optimized formulation is uneconomical, as there is a high likelihood the formulation will not progress to pivotal clinical studies. In order to test safety and efficacy, doses are often up to five to 10 times the estimated therapeutic dose.

Starting doses, however, will be much lower than the estimated therapeutic dose and the dose increases incrementally to ensure safety of volunteers during the study. This is another reason early phase formulations are often very unlike market formulations where an ideal drug product can be designed around the set doses. Phase I formulations need to be flexible in order to cover the wide range of doses expected in the trial, but also to allow flexibility in the doses administered as the clinical protocol is executed. Therefore, a range of doses—for example, 1, 10 and 100 mg—may be developed to allow dosing from 1 mg to 100s of milligrams when multiple units are given.

An additional consideration is that drug solubility is becoming an increasing challenge. Molecules, described as brick dust or grease balls, hamper the bioavailability of the compound, which could have a significant impact on the pharmacokinetic evaluation of Phase I medications. A final potential complication for early phase dosage forms is blinding. Most Phase I clinical trials will be run against placebo, and the formulation will need to be blinded in order to avoid any bias in the reporting of side effects. All of these factors must be taken into consideration when designing a fit-for-purpose formulation for early phase clinical trials (See Figure 1).

API in capsule
In fact, the simplest formulation requires no formulating, as the active is dosed neat, filled into gelatin or HPMC capsules. This is the simplest and fastest route to clinic. The lack of formulation means there is little chance of incompatibility with only the capsule shell as a potential agent, relative to the API stability. If the API is stable, it is most likely that API in capsule will provide a stable formulation, providing suitable packaging is used. To be suitable for API in capsule approach, the dose needs to be within limits of the technology for filling and the maximum amount of API that can fit into a capsule. This leads to a range of 0.1 mg to around 200 mg possible, but these doses will depend on the powder properties of the API.

Manufacture of this type of dosage form can be performed with several different techniques, depending on the dose, such as vacumetric or volume filling. The most common for API in capsule is gravimetric filling, using equipment such as the Xcelodose or Quantos. These techniques individually dispense the API powder into tarred capsules ensuring the correct dose is provided even at the extremes of the ranges. The Xcelodose can even sit within a containment bag, allowing operation under low humidity or inert atmospheres to aid in API stability (see Figure 2).

For doses in excess of API in capsule, or multiples thereof, API-in-bottle may be considered as an approach. In this case, neat API is dosed into bottles for reconstitution at the point of use to be taken as a solution. The principle limitations with this approach are ease of reconstitution and taste. Taste is obviously an issue for patient acceptability, but may also be an issue for blinding of trials, so this is an additional factor that needs to be taken into consideration. Reconstituting the API- in-bottle with a beverage other than water may aid palatability but compatibility with the drink or variability, such as pH in different brands of orange juice, need to be taken into consideration.

In some cases, the API needs to be formulated with excipients. There are two main reasons that products are formulated, one relates to manufacturability and the ability to produce sufficient numbers of material for a trial or the market. For early phase trials, this may not be a major consideration, as patient numbers are low and a technique such as the Xcelodose, which can realistically fill around 300 to 400 capsules per hour, is suitable. For later phase trails, this becomes limiting and formulated dosage forms with more efficient manufacturing methods are used.

The other reason products are formulated is to ensure biopharmaceutical performance of the API. To be suitable for a drug in capsule approach, the API should be freely wettable and soluble. If this is not the case, then poorly dispersed API may be dissolution rate limited, prevent adequate exposure during clinical trials, leading to failure or variability in data.

Blend in capsule
The most common early phase oral formulation is a simple blend in capsule. At its most basic form, this can be just API and filler, although glidants and lubricants may also be added. The purpose of the filler in these formulations is to aid in distributing the API, so it is not presented to the gastrointestinal tract as an agglomerate, thus aiding dissolution. This also bulks up the formulation to aid filling into the capsule. Brim filling, for example, using a Torpac plate, vacumetric or dosator techniques, may be used to fill the capsules, allowing a higher throughput compared to the API in capsule approach. The choice of excipients for this type of formulation is important with flow of the formulation and segregation potential to be considered in design of the formulation. The choice of filler may also have an effect on dissolution rate; therefore, this will need to be tested in a discriminatory method as early as possible.

Another consideration is the potential for chemical incompatibility. Traditionally, this risk could be understood by performing an excipient compatibility study where binary mixtures of API and excipients are stress tested over short periods of time. Because of the simplicity of this type of formulation, it may be preferred to set down prototype formulations rather than binary blends, gaining data on actual formulations and saving time. This approach is recommended where the risk of incompatibility is low, which can form part of an initial risk assessment looking at the API chemistry and API stability data.  Blinding in this type of formulation is relatively easy. For Phase I, opaque white capsules are used most often. If the API is a Swedish orange or brown, a capsule can be used to mask the contents. Therefore, the placebo can simply consist of a filler to a suitable weight.

Powder in bottle
As with API in capsule, where the intended dose is too high to fill into a capsule, a blend formulation can be filled into bottles. Again, the taste of the formulation may be an issue and needs to be taken into consideration. Adding sweeteners or flavors can significantly complicate an early phase formulation, and assessment of the effectiveness of these interventions is difficult without a clinical trial. The bitterness of the API can be assessed during formulation work, using techniques such as the e-tongue2; however, this technique cannot be used to assess the success of the masking agents, but simply as a guide to the requirement for taste masking. The reconstitution of the blend in bottle approach also needs to be considered. If the API is not highly soluble and a suspension, rather than a solution, is expected, then dose homogeneity is a key consideration. For this reason, suspending agents may form part of the formulation to ensure a reproducible suspension can be formed after reconstitution. Although there are additional formulation considerations, a blend in bottle approach can be used for dosing multiple grams of the API.

Tablets
Formulating a tablet for early phase clinical trials is less common because of the added complexity of this type of formulation. In some indications, this type of dosage form may be preferred, for example, oncology or orphan indications. Phase I oncology trials may be carried out in patients, rather than volunteers, and may be extended to allow the patients to continue to receive treatment. For this reason, clinical trial requirements may be larger than in other indications. For orphan drugs, the development pathway may be truncated and; therefore, it may be prudent to start development on a market-like formulation earlier, so product development is not on the critical path. 

Softgels in early phase
Softgel formulations are typically considered for poorly soluble APIs, in order to take advantage of lipid solubility, or as line extensions to marketed drugs.³ However, this formulation type can also be quick and easy in early phase formulations. An API can be delivered in a softgel, as a solution or suspension, allowing a wider range of doses to be delivered. This means early development is performed on filling lines capable of commercial-like output, which would make this one of the easiest formulation types to scale up. Solubility and stability/compatibility are key considerations for this type of formulation, but blinding is easy with the use of opaque capsule shells.

Poorly soluble APIs
If aqueous solubility is very low, conventional fit-for-purpose formulations will not be appropriate. A formulation with the potential to enhance solubility and, therefore, bioavailability, is required. Programs such as Patheon’s Solubility Enhancement Service can be used to rapidly identify which technology is appropriate for a poorly soluble API and prototype formulations produced.

The most common type of commercial formulations for these types of molecule are amorphous dispersions, or lipidic-based formulations.⁴ For this reason, it may be preferred to also look at these types of formulations in early phase. The use of in silico modelling can be used to accelerate early development programs. Once a technology has been selected and shown in vitro to improve solubility, fit-for-purpose versions can be produced for clinic. For example, for a spray dried amorphous system, a blend of spray-dried intermediate with a filler in a capsule may be appropriate for Phase I, with a more conventional formulation developed later on. For a lipidic system a solution, liquid-filled hard shell capsule or Softgel may be used.³

Guiding early phase development
The starting point for developing an early phase formulation, as with any formulation, has to be the drug physicochemical properties and the Target Product Profile (TPP). The TPP at early phase will be necessarily vague—oral formulation, wide dose levels and immediate release, for example. With these inputs, the formulation scientist can recommend a path forward in order to maximize the chance of the molecule in the clinic, while minimizing the development work required to get to a final formulation.