2.2.9 Control Strategy
One of the most significant aspects of ICH M7 [1] is the control section (Section 8). This provides far greater flexibility in terms of strategies to demonstrate absence than are available in the preceding regional guidelines. These are expressed in terms of a series of control options that provide the ability to more widely use chemical/process‐based arguments to assess purging and effectively aligns the guideline with the risk‐based approaches outlined in ICH Q9 [30]. There are four options:
1 Option 1 – include a test for the impurity in the DS specification with an acceptance criterion at or below the acceptable limit.
2 Option 2 – include a test for the impurity in the specification for an input raw material, starting material or intermediate, or as an in‐process control, with an acceptance criterion at or below the acceptable limit.
3 Option 3 – include a test for the impurity in the specification for a raw material, starting material or intermediate, or as an in‐process control, with an acceptance criterion above the acceptable limit of the impurity in the DS. Such a limit is justified based on clear understanding of the fate and purge in question within the downstream process negating the need for any additional testing later in the process.Such an option can be justified when the level of the impurity in the DS will be less than 30% of the acceptable limit by review of data derived from laboratory‐scale spiking experiments.
4 Option 4 – process control. The highly reactive nature of most mutagenic impurities is such that they are unlikely to survive typical downstream processes at levels of concern in the subsequent DS. In order to assess the potential carryover of such impurities requires an understanding of the effect that process conditions have on residual impurity levels. Where there is sufficient confidence that the level of the impurity in the DS will be below the acceptable limit, then no analytical testing may be required and the impurity does not have to be specified.
Examining these options in more detail many would conclude that the ordering should be reversed. As stated above, most mutagenic reagents are highly reactive by nature or deliberate design. They are used in the construct of the DS to form the molecular skeleton of the compound. This high reactivity is the reason why they are also effectively purged. The challenge has been to be able to assess the level of the risk without resorting to extensive analytical testing. Such an approach was developed by Teasdale et al. [29, 31]; this is examined in detail in Chapter 9. In this approach calculations are based on an evaluation of the specific physicochemical properties of the impurity in question (e.g. reactivity/solubility), relative to the downstream processing conditions they will be exposed to. This is based on the scoring system described in Table 2.6.
Table 2.6 Purge values.
| Physicochemical parameters | Purge factor |
|---|---|
| Reactivity | Highly reactive = 100 |
| Moderately reactive = 10 | |
| Low reactivity/unreactive = 1 | |
| Solubility a | Freely soluble = 10 |
| Moderately soluble = 3 | |
| Sparingly soluble = 1 | |
| Volatility | Boiling point >20 °C below that of the reaction/process solvent = 10 |
| Boiling point ± 10 °C that of the reaction/process solvent = 3 | |
| Boiling point >20 °C above that of the reaction/process solvent = 1 | |
| Ionizability | Ionizaion potential of genotoxic impurity (GI) significantly different to that of the desired productb |
| Physical processes – chromatography | Chromatography – GI elutes prior to desired product = 100 |
| Chromatography – GI elutes after desired product = 10 | |
| Others evaluated on an individual basis |
a This relates to solubility within the context of a recrystallization/isolation process whereby the impurity in question, if highly soluble, will remain within mother liquors and hence be purged from the desired product.
b This relates to a deliberate attempt to partition the desired product/GI between an aqueous and organic layer, typically achieved through the manipulation of pH to change the ionized/unionized state of one of the components.
This specific approach is referenced directly within the guideline and its application examined in detail in the attached case study.
Examples of the options described are further examined within the guideline through a series of case studies. There are four cases in total: case studies 1 and 2 examine the use of Option 3, case study 3 examines the use of Options 2 and 4 combined, and case study 4 examines the use of Option 4.
Case study 3 (guideline) – reproduced below, is concerning.
The Step 1 intermediate of a five‐step synthesis is a nitroaromatic compound that may contain low levels of impurity C, a positional isomer of the Step 1 intermediate and also a nitroaromatic compound. The amount of impurity C in the Step 1 intermediate has not been detected by ordinary analytical methods, but it may be present at lower levels. The Step 1 intermediate is positive in the bacterial mutagenicity assay. The Step 2 hydrogenation reaction results in a 99% conversion of the Step 1 intermediate to the corresponding aromatic amine. This is confirmed via in‐process testing. An assessment of purge of the remaining Step 1 nitroaromatic intermediate was conducted, and a high purge factor was predicted based on purge points in the subsequent Steps 3 and 4 processing steps. Purge across the Step 5 processing step is not expected, and a specification for the Step 1 intermediate at the TTC‐based limit was established at the Step 4 intermediate (Option 2 control approach). The positional isomer impurity C would be expected to purge via the same purge points as the Step 1 intermediate and therefore will always be much lower than the Step 1 intermediate itself and therefore no testing is required and an Option 4 control strategy for impurity C can be supported without the need for any additional laboratory or pilot scale data.
In this example during Step 1 the nitro‐aromatic is reported to be converted at high yield (99%) conversion to an aromatic amine; further reductions in level are reported to occur at Steps 3 and 4. Certainly, were such a scenario encountered in reality, it would seem logical to conduct a risk assessment based on an evaluation of the purging capacity of the process before simply defaulting to a control strategy based on Option 2.
2.2.9.1 Considerations for Control Approaches
Within the guideline, this subsection of the overall control strategy section provides further clarification as to what is required to support the specific control options in terms of data. Of particular note is the need to take into consideration the effect of scale, i.e. it is important to address the expected scale dependence or independence of any data especially where supporting data are derived from lab or pilot scale manufacture.
If options 3 and 4 cannot be justified, then a test for the impurity on the specification