The present invention is concerned with a method of monitoring polymorph purity, in particular in relation to erlotinib, and compositions obtained further to such monitoring.

Erlotinib, N-(3-ethynylphenyl)-6,7-bis(2-methoxyethoxy)-4-quinazolinami ne, has the following structural Formula

WO 01/34574 discloses erlotinib hydrochloride Forms A and B. WO 01/34574 teaches that Form A is thermodynamically less stable than Form B. In particular, page 16 of WO 01/34574 discloses that polymorph B of erlotinib hydrochloride was found to be the thermodynamically most stable and desirable Form. On this basis, WO 01/34574 aimed to provide Form B in the substantially pure polymorphic Form, and also pharmaceutical compositions of the substantially pure Form B, particularly as a tablet and a method of the selective production of the compound. Thus although Form A is disclosed in WO 01/34575, the main aim of this document was to disclose pure Form B and compositions including the same.

Also according to WO 01/34574, the disclosure of the original compound patent for erlotinib, namely US 5747498, actually disclosed the provision of a mixture of the polymorphs Form A and B. Still further according to WO 01/34574, because of the hydrochloride polymorph reduced stability, in particular due to the thermodynamically less stable Form A component, the hydrochloride salt was not more preferred for tablet formulation than the mesylate salt forms.

Specifically according to WO 01/34574, therefore, there was disclosed substantially homogeneous erlotinib hydrochloride Form B that exhibits an X-ray powder diffraction pattern having characteristic peaks expressed in degrees 2-theta at approximately 6.26, 12.48, 13.39, 16.96, 20.20, 21.10, 22.98, 24.46, 25.14 and 26.91. Form B was also characterized in WO 01/34574 by the X-ray powder diffraction pattern shown in Figure 3 of WO 01/34574. WO 01/34574 also disclosed erlotinib hydrochloride Form B that exhibits an X-ray powder diffraction pattern having characteristic peaks expressed in degrees 2-theta at approximately 6.26, 12.48, 13.39, 16.96, 20.20, 21.10, 22.98, 24.46, 25.14 and, 26.91, which is substantially free of the erlotinib hydrochloride polymorph designated the A polymorph.

Erlotinib hydrochloride Form B is in particular characterized by the above peak as disclosed in WO 01/34574 at 6.26 degrees 2-theta, although it can be appreciated that there is some measurement tolerance with different X-ray powder diffraction apparatus and conditions.

Also according to WO 01/34574, there is disclosed a pharmaceutical composition for the treatment of a hyperproliferative disorder in a mammal which substantially comprises a therapeutically effective amount of the polymorph Form B and a pharmaceutically acceptable carrier. The pharmaceutical composition may be adapted for oral administration. It may be in the form of a tablet.

Still further according to WO 01/34574, there is disclosed a composition consisting essentially of erlotinib hydrochloride polymorph Form A, which is characterized by an X-ray powder diffraction pattern shown in Figure 1 of WO 01/34574. Form A is further characterized in WO 01/34574 by the peaks shown in Table 1 or Table 2 of WO 01/34574 as below.

The time per step for the above analysis is 1.0 second.

What is particularly noticeable from the above, is the low intensity peak given at 6.165 degrees 2-theta in WO 01/34574 as being part of the 2-theta characterization allegedly attributable to Form A. This 6.165 peak is, however, indicative of Form B contamination in the Form A sample of WO 01/34574. Such contamination is not surprising given the thermodynamic instability associated with Form A as taught by WO 01/34574.

WO 01/34574 also discloses the following characterization of Form A, namely an X-ray powder diffraction pattern having characteristic peaks expressed in degrees 2-theta at approximately 5.58, 9.84, 11.25, 18.86, 22.70, 23.50, 24.18, 24.59, 25.40 and 29.24.

WO 2009/024989 reports that the erlotinib hydrochloride crystalline polymorph Form A obtained by the processes described in the art, for example by the processes described and exemplified in the US 5747498 and WO 01/34574, were contaminated with polymorph Form B. Specifically, the experimental data disclosed in WO 01/34574 showed that the polymorph Form A had a peak in an X-ray powder diffraction pattern at about 6.26 ± 0.2 degrees 2-theta, which is the characteristic peak of polymorph Form B as discussed above. This WO 2009/024989 analysis agrees with our interpretation of WO 01/34574 as provided above.

According to WO 2009/024989, therefore, there was a need in the art for a process for producing erlotinib hydrochloride crystalline polymorph Form A substantially free of polymorph Form B. Thus, there was allegedly provided by WO 2009/024989 erlotinib hydrochloride crystalline polymorph Form A substantially free of polymorph Form B, and specifically erlotinib hydrochloride crystalline polymorph Form A characterized by peaks in the powder X-ray diffraction pattern having 2-theta angle positions at about 5.75, 9.88, 11.40, 18.97, 22.84, 23.65, 24.29, 24.75, 25.56 and 29.37 ± 0.2 degrees 2-theta and by the absence of a peak at about 6.26 ± 0.2 degrees 2-theta.

Also according to WO 2009/024989, there was provided a pharmaceutical composition comprising erlotinib hydrochloride crystalline polymorph Form A characterized by peaks in the powder X-ray diffraction pattern having 2-theta angle positions at about 5.75, 9.88, 11.40, 18.97, 22.84, 23.65, 24.29, 24.75, 25.56 and 29.37 ± 0.2 degrees and by the absence of a peak at about 6.26 ± 0.2 degrees 2-theta. WO 2009/024989 further describes that the term "eriotinib hydrochloride crystalline polymorph Form A substantially free of polymorph Form B" refers to the eriotinib hydrochloride polymorph Form A containing less than about 10% crystalline polymorph Form B of eriotinib hydrochloride, preferably less than 5% crystalline polymorph Form B of eriotinib hydrochloride, more preferably less than 1% crystalline polymorph Form B of eriotinib hydrochloride, and still more preferably essentially free of crystalline polymorph Form B of eriotinib hydrochloride. "Essentially free of crystalline polymorph Form B of eriotinib hydrochloride" means that no crystalline polymorph Form B of eriotinib hydrochloride can be detected within the limits of a powder X-ray diffractometer. However, such statement in WO 2009/024989 can only be verified in practice by the authors based on the specific apparatus and associated operating conditions from WO 2009/024989.

WO 2009/024989 further describes that the X-ray powder diffraction spectrum as referenced therein was measured on a bruker axs D8 advance X-ray powder diffractometer having a copper-ka radiation. Approximately 1 gm of sample was gently flattened on a sample holder and scanned from 2 to 50 degrees two-theta, at 0.03 degrees two-theta per step and a time per step of 0.5 seconds. The sample was simply placed on the sample holder. The sample was rotated at 30 rpm at a voltage 40KV and 35 mA. However, when one looks at the specific X-ray powder diffraction analysis provided in WO 2009/024989, this can be seen to be of unacceptable sensitivity and in particular the noise level is too high for the parameters used. Thus, there is no enabling disclosure in WO 2009/024989 as to the provision of pure eriotinib hydrochloride Form A.

Further to the above discussion, therefore, we have now discovered that while both WO 01/34574 and WO 2009/024989 allege to provide eriotinib hydrochloride Form A, neither processes of preparation, nor X-ray powder diffraction techniques, meaningfully and reliably enable the provision of sufficiently pure eriotinib hydrochloride Form A, whereby a composition including the same as an input API would have the required stability profile on storage and / or dissolution profile. We have, therefore, now developed a pharmaceutical composition comprising eriotinib hydrochloride, together with one or more pharmaceutically acceptable carriers, wherein the eriotinib hydrochloride is present as Form A and which composition has acceptable storage and dissolution properties. In particular, the compositions of the invention are essentially free of eriotinib hydrochloride Form B, when analysed by the X-ray powder diffraction techniques as described herein.

According to the present invention, therefore, there is provided eriotinib hydrochloride Form A that includes no detectable eriotinib hydrochloride Form B, as determined by X-ray powder diffraction analysis that includes a differentiation resolution and a 2-theta scan region that enables detection of the absence, or presence, of a characteristic 2-theta peak of eriotinib hydrochloride Form B in the region of 6.0 to 6.4 degrees 2-theta.

Preferably, the above X-ray powder diffraction analysis includes a differentiation resolution between adjacent 2-theta peaks of not more than about 0.1 degrees 2-theta. Still further, it is preferred that the above X-ray powder diffraction analysis has a detection limit that enables detection of the presence of equal to, or less than, 0.2% eriotinib hydrochloride Form B in a test sample.

An example of suitable X-ray powder diffraction apparatus that enables the above level of detection is a Bruker AXS D8, with a radiation source of Cu Κα (λ=1.54 A) and a detector type LynxEye. Still further for the above analysis, there are key operating parameters that need to be balanced to enable the sensitive analysis required by the present invention. These parameters include current, voltage, step size and time per step, and each of these parameters should be individually controlled and correlated to each of the other operating parameters as above. One example of selected parameters to enable the level of detection includes a current of 40 mA, a voltage of 40 kV, a step size of 0.02 degrees 2-theta and a time per step of 4.0 seconds.

Still further, it is preferred that the above X-ray powder diffraction is operated with a divergence slit of 0.6 mm. The preferred start angle is 4 degrees 2-theta, and the preferred end angle is 7 degrees 2-theta. The diffractometer is operated in a continuous scan mode. The typical sample holder is a PMMA holder with a 25 mm diameter and 1.5 mm depth circular cavity for spreading the sample, with 25 x 75 x 1 mm glass slides. Typically, the sample is prepared by spreading the powder in the sample holder with the help of a glass slide.

There is also provided by the present invention a method of monitoring polymorphic purity of an eriotinib hydrochloride sample, the method comprising an X-ray powder diffraction analysis that includes a differentiation resolution and a 2-theta scan region that enables detection of the absence, or presence, of a characteristic 2-theta peak of eriotinib hydrochloride Form B in the region of 6.0 to 6.4 degrees 2-theta.

It is further preferred that the above X-ray powder diffraction analysis includes a differentiation resolution between adjacent 2-theta peaks of not more than about 0.1 degrees 2- theta. Still further, it is preferred that the above X-ray powder diffraction analysis has a detection limit that enables detection of the presence of equal to, or less than, 0.2% eriotinib hydrochloride Form B in a test sample.

In a preferred embodiment, the method employs a Bruker AXS D8, with a rad iation source of Cu Κα (λ=1.54 A) and a detector type LynxEye. Still further, this is operated with a current of 40 mA, a voltage of 40 kV, a step size of 0.02 degrees 2-theta and a time per step of 4.0 seconds.

The above analysis can be carried out at any stage of the development process, such as analysis of the API per se; a formulation mixture to be granulated; the granules themselves; and / or still further a final dosage form, such as a tablet core prior to coating and / or the resulting coated dosage form.

The present invention further provides a solid pharmaceutical composition, comprising eriotinib hydrochloride Form A, together with one or more pharmaceutically acceptable excipients, which composition includes eriotinib hydrochloride Form A as hereinbefore described as the input API. I n other words, the composition is prepared from eriotinib hydrochloride Form A as hereinbefore described.

There is also provided by the present invention a solid pharmaceutical composition, comprising eriotinib hydrochloride Form A, together with one or more pharmaceutically acceptable excipients, whereby such composition exhibits a stability profile whereby on storage at (i) a temperature of at least about 23°C, (ii) a relative humidity of at least about 55%, and (iii) for a period of at least about 6 months, no eriotinib hydrochloride Form B is detectable by X-ray powder diffraction analysis. For example, no characteristic eriotinib hydrochloride Form B 2-theta peak in the region of 6.0 to 6.4 degrees 2-theta is detectable by X-ray powder diffraction analysis. Preferably, the X-ray powder diffraction analysis is as herein described on a Bruker AXS D8 and with operating parameters also as herein described. Such stability profile is provided as a result of an input eriotinib hydrochloride API being of sufficient Form A purity, for example as determined by X-ray powder diffraction analysis as herein described.

Even more preferably, the present invention provides a solid pharmaceutical composition, comprising eriotinib hydrochloride Form A, together with one or more pharmaceutically acceptable excipients, whereby such composition exhibits a stability profile whereby on storage at (i) a temperature of at least about 38°C, (ii) a relative humidity of at least about 70%, and (iii) for a period of at least about 6 months, no erlotinib hydrochloride Form B is detectable by X-ray powder diffraction analysis. For example, no characteristic erlotinib hydrochloride Form B 2-theta peak in the region of 6.0 to 6.4 degrees 2-theta is detectable by X-ray powder diffraction analysis. Preferably, the X-ray powder diffraction analysis is as herein described on a Bruker AXS D8 and with operating parameters also as herein described. Again, such stability profile is provided as a result of an input erlotinib hydrochloride API being of sufficient Form A purity, for example as determined by X-ray powder diffraction analysis as herein described.

The above stability profile on storage can be supported, for example, by reference to Figures 1 to 4, that illustrate the absence of a characteristic erlotinib hydrochloride Form B 2-theta peak in the region of 6.0 to 6.4 degrees 2-theta detectable by X-ray powder diffraction analysis under accelerated storage conditions as above. The specific composition analysed in these Figures is a 150 mg compositions as set out in Example 9. The X-ray powder diffraction analysis of the tablets indicate that no transformation has occurred following 6 months of storage under accelerated conditions and pure polymorph Form A has been maintained.

As indicated above, typically such a stability profile is provided as a result of an input erlotinib hydrochloride API being of sufficient Form A purity, for example as determined by X-ray powder diffraction analysis according to the present invention. Typically, such input API has a slow scan X-ray powder diffraction pattern showing no traces of Form B present as shown in Figure 6 and with further reference to Example 1 hereinafter.

The importance of pure input API can be further illustrated by reference to Example 2, and associated Figures 7 to 12. Tablets (25 mg and 150 mg compositions as set out in Example 9) were stored at 40°C ± 2°C and 75% ± 5% relative humidity for a period of 3 months. However, the input API included trace amounts of erlotinib hydrochloride Form B according to Example 2. The X-ray powder diffraction pattern of the tablets in Figures 7 to 12 indicate transformation to Form B has occurred for both lowest and highest strength (25mg and 150mg) following 3 months of storage under accelerated conditions.

It should be pointed out that the provided results for the stability measurements were generated in the Examples based on a conventional X-ray powder diffraction scan and therefore the peak at 6.2 ± 0.2 degrees 2-theta was absent from the initial analysis results. Traces of Form B for both strengths were however then identified by visual observation of the X-ray powder diffraction pattern whereby a minute peak appears at the 2-theta value of 6.2 ± 0.2 on storage. Further to such observation, a slow scan X-ray powder diffraction was carried out (Example 2) for the API used for the manufacturing of the stability batches and indicated that traces of Form B were originally present in the API.

According to a further aspect of the present invention, there is provided a solid pharmaceutical composition, comprising eriotinib hydrochloride Form A, together with one or more selected pharmaceutically acceptable excipients, whereby such composition exhibits a dissolution profile of at least 40% at 30 minutes when measured in 1000 m L volume of 0.1N hydrochloric acid, at a temperature of 37°C and a paddle speed of 75 rpm. The dissolution is measured using USP-II apparatus (Paddle). Such dissolution profile is provided at least as a result of an input eriotinib hydrochloride API being of sufficient Form A purity, for example as determined by X-ray powder diffraction analysis as herein described. Preferably, the X-ray powder diffraction analysis is as herein described on a Bruker AXS D8 and with operating parameters also as herein described.

According to a further aspect of the present invention, there is provided eriotinib hydrochloride Form A, that includes no detectable eriotinib hydrochloride Form B by X-ray powder diffraction analysis as herein described, wherein said eriotinib hydrochloride Form A has a dgo below about 25 microns, when measured with a Malvern Mastersizer 2000. More preferably, the dgo is in the range of about 5 to 15 microns, and even more preferably in the range of about 9 to 14 microns.

The PSD of eriotinib hydrochloride Form A as provided by the present invention can be still further characterized by a dso in the range of about 1 to 6 microns, when measured with a Malvern Mastersizer 2000, and more preferably in the range of about 2 to 5 microns.

Still further, the PSD of eriotinib hydrochloride Form A as provided by the present invention can be further characterized by a dio in the range of about 0.5 to 2.5 microns, when measured with a Malvern Mastersizer 2000, and more preferably in the range of about 0.5 to 1.5 microns.

There is further provided by the present invention eriotinib hydrochloride Form A, that includes no detectable eriotinib hydrochloride Form B when measured by X-ray powder diffraction analysis as herein described, and which is further characterized by a PSD as shown in Figure 25 as measured by a Malvern Mastersizer 2000. There is further provided by the present invention eriotinib hydrochloride Form A, that includes no detectable eriotinib hydrochloride Form B when measured by X-ray powder diffraction analysis as herein described, and which is further characterized by a PSD as shown in Figure 26 as measured by a Malvern Mastersizer 2000.

The specific operating parameters of the Malvern Mastersizer 2000 as used in the measurements of the present invention are provided hereinafter in greater detail in the Experimental portion of the description.

It can be further understood that according to a particularly preferred aspect of the present invention eriotinib hydrochloride and associated compositions are provided having more than one, preferably all, of the preferred invention features as described herein, namely purity dissolution profile, stability profile and / or particle size. This overall provision of the present invention is further illustrated in the claims of the present application.

According to a further aspect of the present invention, this further provides a process for preparing a pharmaceutical composition comprising eriotinib hydrochloride Form A, and at least one pharmaceutically acceptable carrier, comprising: obtaining a batch of eriotinib hydrochloride Form A; analyzing the batch for the presence of eriotinib hydrochloride Form B as described herein; preparing a pharmaceutical composition from the batch only if the batch is determined to have the required polymorphic purity of eriotinib hydrochloride Form A, whereby there is no detectable eriotinib hydrochloride Form B as described herein. The determination of polymorphic purity and the confirmation of the absence of Form B, is carried out by the X-ray powder diffraction analysis of the invention as described herein.

According to a still further aspect of the present invention, this further provides a process for preparing a packaged pharmaceutical composition comprising eriotinib hydrochloride Form A, comprising: obtaining a pharmaceutical composition comprising eriotinib hydrochloride Form A; analyzing the pharmaceutical composition for the presence of eriotinib hydrochloride Form B as described herein; and packaging the pharmaceutical composition only if the eriotinib hydrochloride is determined to have the required polymorphic purity of eriotinib hydrochloride Form A, whereby there is no detectable eriotinib hydrochloride Form B as described herein. The determination of polymorphic purity and the confirmation of the absence of Form B, is carried out by the X-ray powder diffraction analysis of the invention as described herein.

According to a still further aspect of the present invention, this further provides a process of distributing a validated batch of a pharmaceutical composition comprising eriotinib hydrochloride Form A and at least one pharmaceutically acceptable carrier, comprising: obtaining a batch of the pharmaceutical composition; performing stability testing with a sample of the batch; analyzing the sample of the batch for the presence of eriotinib hydrochloride Form B as described herein after said stability testing; validating the batch for distribution only if the sample of the batch after stability testing is determined to have the required polymorphic purity of eriotinib hydrochloride Form A, whereby there is no detectable eriotinib hydrochloride Form B as described herein (the determination of polymorphic purity and the confirmation of the absence of Form B, is carried out by the X-ray powder diffraction analysis of the invention as described herein); and distributing the validated batch.

The one or more excipients as used in a solid oral dosage form according to the present invention preferably include pharmaceutically acceptable excipients, such as fillers, binders, disintegrants, glidants, lubricants and the like. Suitable fillers for inclusion in solid oral dosage forms according to the present invention include sugars, sugar alcohols and polymeric glycosides. Examples of sugars are sucrose, glucose and lactose as the monohydrate or in anhydrous form. Examples of sugar alcohols include mannitol, xylitol and sorbitol. Examples of polymeric glycosides are maltodextrin, microcrystalline cellulose and starches of different origins. Particularly suitable fillers include lactose and / or microcrystalline cellulose or the like.

Suitable binders for inclusion in solid oral dosage forms according to the present invention so as to ensure the required mechanical strength, include wet and / or dry binders depending on the formulation process employed. Typical binders include polymers, such as polyvinylpyrrolidone, and cellulose derivatives, such as hydroxypropyl cellulose and microcrystalline cellulose. In certain embodiments, a separate binder might not be necessary, and sufficient binding function might be provided by the dual excipient function of a filler excipient, such as microcrystalline cellulose.

Examples of suitable disintegrants for inclusion in solid oral dosage forms according to the present invention are crospovidone and croscarmellose, such as croscarmellose sodium, starches and modified starches, e.g. maize starch, in pregelatinized form or as sodium glycolate, and hydroxypropyl cellulose with a low degree of substitution (L-H PC). A preferred disintegrant is sodium starch glycolate.

Examples of suitable glidants, include colloidal silicon dioxide, calcium silicates and talcum.

Preferred lubricants include stearic acid or salts thereof, with a particularly preferred lubricant comprising magnesium stearate.

It will be further understood that compositions according to the present invention as described herein can be used in the treatment of conditions that can be alleviated by the administration of eriotinib. Specifically, eriotinib is indicated for the treatment of non-small cell lung cancer or pancreatic cancer. Accordingly, there is provided by the present invention eriotinib hydrochloride Form A as described herein, for the treatment of cancer.

The following Figures, Experimental and Examples are provided to illustrate the invention and are not to be construed as limiting the scope of the invention in any manner. Figure la: X-ray powder diffraction pattern (raw data) of 150mg tablet according to Example 9 at time zero. ["Pure" Form A input API - Example 1]

Figure lb: X-ray powder diffraction pattern (peak search) of 150mg tablet according to Example 9 at time zero. ["Pure" Form A input API - Example 1]

Figure lc: Peak search results from Figure lb. ["Pure" Form A input API - Example 1]

Figure 2: X-ray powder diffraction pattern (raw data) of 150mg tablet according to Example 9 at 6 months storage at accelerated conditions, 40°C ± 2°C and 75% ± 5% relative humidity. ["Pure" Form A input API - Example 1]

Figure 3: X-ray powder diffraction pattern (peak search) of 150mg tablet according to Example 9 at 6 months storage at accelerated conditions, 40°C ± 2°C and 75% ± 5% relative humidity. ["Pure" Form A input API - Example 1]

Figure 4: Peak search results from Figure 3. ["Pure" Form A input API - Example 1]

Figure 5: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 1 ["Pure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 6: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 1 ["Pure" Form A], the 2-theta region from 4-7 was scanned with a 20 times slower scan rate compared to Figure 5.

Figure 7a: X-ray powder diffraction pattern (raw data) of 150mg tablet according to Example 9 at time zero. ["Impure" Form A input API - Example 2]

Figure 7b: X-ray powder diffraction pattern (peak search) of 150mg tablet according to Example 9 at time zero. ["Impure" Form A input API - Example 2]

Figure 7c: Peak search results from Figure 7b. ["Impure" Form A input API - Example 2] Figure 8: X-ray powder diffraction pattern (raw data) of 150mg tablet according to Example 9 at 3 months storage at accelerated conditions, 40°C ± 2°C and 75% ± 5% relative humidity. ["Impure" Form A input API - Example 2]

Figure 9: X-ray powder diffraction pattern (peak search) of 150mg tablet according to Example 9 at 3 months storage at accelerated conditions, 40°C ± 2°C and 75% ± 5% relative humidity. ["Impure" Form A input API - Example 2]

Figure 10: Peak search results from Figure 9. ["Impure" Form A input API - Example 2]

Figure 11: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 2 ["Impure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 12: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 2 ["Impure" Form A], the 2-theta region from 4-7 was scanned with a 20 times slower scan rate compared to Figure 11.

Figure 13: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 3 ["Pure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 14: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 3 ["Pure" Form A], the 2-theta region from 4-7 was scanned with a 13 times slower scan rate compared to Figure 13.

Figure 15: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 4 [present in a dry mixture] ["Pure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 16: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 4 [present in a dry mixture] ["Pure" Form A], the 2-theta region from 4-7 was scanned with a 13 times slower scan rate compared to Figure 15. Figure 17: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 5 [present in granules] ["Pure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 18: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 5 [present in granules] ["Pure" Form A], the 2-theta region from 4-7 was scanned with a 13 times slower scan rate compared to Figure 17.

Figure 19: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 6 ["Impure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 20: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 6 ["Impure" Form A], the 2-theta region from 4-7 was scanned with a 13 times slower scan rate compared to Figure 19.

Figure 21: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 7 [present in a dry mixture] ["Impure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 22: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 7 [present in a dry mixture] ["Impure" Form A], the 2-theta region from 4-7 was scanned with a 13 times slower scan rate compared to Figure 21.

Figure 23: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 8 [present in granules] ["Impure" Form A], and digitized patterns of eriotinib hydrochloride Forms A and B according to WO 01/34574.

Figure 24: X-ray powder diffraction pattern of eriotinib hydrochloride of Example 8 [present in granules] ["Impure" Form A], the 2-theta region from 4-7 was scanned with a 13 times slower scan rate compared to Figure 23.

Figure 25: A PSD of eriotinib hydrochloride Form A. Figure 26: A PSD of erlotinib hydrochloride Form A. Experimental:

X-ray Powder Diffractometer:

Instrument type used: Bruker AXS D8

Source: Cu Κα (λ=1.54°Α)

Detector type: LynxEye

Divergence slit: 0.6 mm

Voltage: 40kV

Current: 40mA

Background removal: Yes using DIFF AC plus EVA 12.0 software

Start angle: 3 2-theta for reference analysis, 4 2-theta for scan using conditions of the invention

End angle: 35 2-theta for reference analysis, 7 2-theta for scan using conditions of the invention

Step size: 0.02 2-theta

Scanning rate: 0.5 s / step for reference analysis, 4 s / step for scan using conditions of the invention

Scan type: Locked Coupled

Scan mode: Continuous

Sample holder: PMMA holder with a 25 mm diameter and 1.5 mm depth circular cavity for spreading the sample.

Glass slides: 25x75x1 mm glass slides

Sample Preparation: The sample was prepared by spreading the powder in sample holder with the help of a glass slide.

Malvern Mastersizer:

Principle:

Laser Diffraction technique

Wet dispersion method

Equipment: Malvern Mastersizer 2000 Sample Handling Unit: Hydro 2000 S (A)

Instrument Parameters:

Material Refractive Index: 1.533

Material absorption: 0.1

Dispersion medium: Buffer pH 7.4

Dispersant refractive index: 1.330

Models: General purpose

Calculation Sensitivity: Enhanced

Sample measurement: Measurement time 12 seconds, background time 10 seconds

Obscuration limits: 10 to 20%

Stirrer speed: 2500 rpm

Internal ultrasound: 70%

Measurement cycles: Aliquot 1, measurement 3 per aliquot, calculate the average

Apparatus: Analytical balance, pipettes, beakers

Reagent: Water, monopotassium phosphate, sodium hydroxide, phosphoric acid, Tween 80 Example 1:

Eriotinib hydrochloride was analyzed by X-ray powder diffraction. The X-ray powder diffraction pattern along with the digitized patterns of eriotinib hydrochloride Forms A and B of WO 01/34574, can be found in Figure 5. The reflections are also set out in the following table.

16.921 5.23544 459 1.5

17.940 4.94057 213 0.7

18.413 4.81467 224 0.7

18.881 4.69635 2994 9.9

19.569 4.53276 589 1.9

20.608 4.30650 404 1.3

21.885 4.05799 369 1.2

22.768 3.90264 1725 5.7

23.535 3.77707 5721 18.9

24.225 3.67096 1838 6.1

24.550 3.62317 1484 4.9

25.451 3.49684 1414 4.7

26.210 3.39732 1272 4.2

29.258 3.05000 1071 3.5

30.047 2.97168 311 1.0

30.691 2.91080 391 1.3

Based on the above data and the polymorph characterization of WO 01/34574, it is understood that the polymorph of Example 1 is eriotinib hydrochloride Form A. In order to confirm the absence of even trace amounts of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 20 times slower scan rate and the results are shown in Figure 6. From Figure 6 it is apparent that the characteristic polymorph B peak at 6.3 2-theta is missing. This confirms that the polymorph of Example 1 is pure eriotinib hydrochloride Form A with not even trace amounts of Form B.

Example 2:

Eriotinib hydrochloride was analyzed by X-ray powder diffraction. The X-ray powder diffraction pattern along with the digitized patterns of eriotinib hydrochloride Forms A and B of WO 01/34574, can be found in Figure 11. The reflections are also set out in the following table.

30.072 2.96921 373 0.8

30.700 2.90996 398 0.8

In order to confirm the absence / presence of even trace amounts of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 20 times slower scan rate and the results are shown in Figure 12. From Figure 12 it is apparent that the characteristic polymorph B peak at 6.3 2- theta is present. This confirms that the polymorph of Example 2 is eriotinib hydrochloride Form A in admixture with Form B.

Example 3:

Eriotinib hydrochloride was analyzed by X-Ray powder diffraction. The X-Ray powder diffraction pattern along with the digitized patterns of polymorphs A and B of eriotinib hydrochloride Forms A and B of WO 01/34574, can be found in Figure 13. The reflections are also set out in the following table.

20.686 4.29029 349 0.9

21.961 4.04411 348 0.9

22.826 3.89270 1728 4.3

23.597 3.76726 6325 15.9

24.303 3.65944 1965 4.9

24.608 3.61479 1509 3.8

24.997 3.55943 1321 3.3

25.507 3.48938 1468 3.7

25.885 3.43928 573 1.4

26.297 3.38635 1322 3.3

26.677 3.33888 636 1.6

27.124 3.28493 429 1.1

27.366 3.25634 351 0.9

29.303 3.04541 1192 3.0

30.126 2.96409 294 0.7

30.764 2.90398 384 1.0

31.411 2.84565 226 0.6

31.827 2.80945 546 1.4

32.848 2.72437 478 1.2

Based on the above data and the polymorph characterization of WO 01/34574, it is understood that the polymorph of Example 3 is eriotinib hydrochloride Form A. In order to confirm the absence of even trace amounts of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 13 times slower scan rate and the results are shown in Figure 14. From Figure 14 it is apparent that the characteristic polymorph B peak at 6.3 2-theta is missing. This confirms that the polymorph of Example 3 is pure eriotinib hydrochloride Form A with not even trace amounts of Form B.

Example 4:

Eriotinib hydrochloride dry mixture was analyzed by X-Ray powder diffraction. The X-Ray powder diffraction pattern along with the digitized patterns of polymorphs A and B of eriotinib hydrochloride Forms A and B of WO 01/34574, and the placebo pattern, can be found in Figure 15. The reflections are also set out in the following table.

26.584 3.35037 2150 2.6

27.034 3.29561 1260 1.5

27.448 3.24686 1590 1.9

28.192 3.16281 803 1.0

28.498 3.12958 1017 1.2

Based on the above data and the polymorph characterization of WO 01/34574, it is understood that the polymorph of the dry mixture of Example 4 is eriotinib hydrochloride Form A. In order to confirm the absence of even trace amounts of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 13 times slower scan rate and the results are shown in Figure 16. From Figure 16 it is apparent that the characteristic polymorph B peak at 6.3 2-theta is missing. This confirms that the polymorph of Example 4 is pure eriotinib hydrochloride Form A with not even trace amounts of Form B.

Example 5:

Eriotinib hydrochloride granules was analyzed by X-Ray powder diffraction. The X-Ray powder diffraction pattern along with the digitized patterns of polymorphs A and B of eriotinib hydrochloride Forms A and B of WO 01/34574, and the placebo pattern, can be found in Figure 17. The reflections are also set out in the following table.

10.301 8.58083 2841 11.0

11.260 7.85186 2729 10.5

11.750 7.52530 1834 7.1

12.433 7.11345 7817 30.2

12.747 6.93930 1919 7.4

13.131 6.73684 1781 6.9

13.469 6.56859 1684 6.5

14.329 6.17633 1339 5.2

14.690 6.02523 1048 4.0

15.150 5.84336 898 3.5

15.588 5.68027 740 2.9

16.313 5.42925 9948 38.4

16.974 5.21947 2117 8.2

18.908 4.68963 9721 37.5

19.064 4.65166 14216 54.9

19.525 4.54273 16167 62.5

19.925 4.45258 25888 100.0

20.759 4.27548 6181 23.9

21.185 4.19038 12766 49.3

21.970 4.04249 2417 9.3

22.723 3.91013 9215 35.6

23.524 3.77884 18196 70.3

24.216 3.67241 5032 19.4

24.536 3.62523 4304 16.6

24.929 3.56894 3880 15.0

25.501 3.49015 7802 30.1

26.172 3.40212 4178 16.1

26.512 3.35930 2119 8.2

26.780 3.32632 1698 6.6

Based on the above data and the polymorph characterization of WO 01/34574, it is understood that the polymorph of the granules of Example 5 is eriotinib hydrochloride Form A. In order to confirm the absence of even trace amounts of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 13 times slower scan rate and the results are shown in Figure 18. From Figure 18 it is apparent that the characteristic polymorph B peak at 6.3 2-theta is missing. This confirms that the polymorph of Example 4 is pure eriotinib hydrochloride Form A with not even trace amounts of Form B.

Example 6:

Eriotinib hydrochloride was analyzed by X-Ray powder diffraction. The X-Ray powde diffraction pattern along with the digitized patterns of polymorphs A and B of eriotinib hydrochlorid Forms A and B of WO 01/34574, can be found in Figure 19. The reflections are also set out in th following table.

20.627 4.30258 484 0.9

21.963 4.04381 378 0.7

22.804 3.89650 2039 3.8

23.560 3.77308 6201 11.5

24.260 3.66585 1738 3.2

24.571 3.62005 1347 2.5

25.445 3.49772 1290 2.4

26.238 3.39382 1290 2.4

29.278 3.04796 1154 2.1

30.089 2.96758 317 0.6

30.711 2.90888 409 0.8

31.536 2.83471 296 0.6

31.797 2.81201 557 1.0

32.837 2.72529 545 1.0

34.508 2.59705 170 0.3

Based on the above data and the polymorph characterization of WO 01/34574, it is understood that the polymorph of Example 6 is eriotinib hydrochloride Form A. In order to detect possible presence of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 13 times slower scan rate and the results are shown in Figure 20. From Figure 20, it is apparent that the characteristic polymorph B peak at 6.3 2-theta is present so API is mainly polymorph A, but presence of minor quantity of Form B (approx. 0.5 %) was detected.

Example 7:

Eriotinib hydrochloride dry mixture was analyzed by X-Ray powder diffraction. The X-Ray powder diffraction pattern along with the digitized patterns of polymorphs A and B of eriotinib hydrochloride Forms A and B of WO 01/34574, and the placebo pattern, can be found in Figure 21. The reflections are also set out in the following table. Angle d value Intensity Intensity % 2-Theta ° Angstrom Count %

5.638 15.66249 43252 75.6

6.181 14.28666 371 0.6

9.796 9.02164 4723 8.3

10.358 8.53382 2469 4.3

11.316 7.81348 3056 5.3

11.794 7.49738 1514 2.6

12.479 7.08774 9141 16.0

13.510 6.54881 1279 2.2

14.682 6.02841 574 1.0

16.350 5.41706 4568 8.0

17.017 5.20634 1775 3.1

18.908 4.68963 9452 16.5

19.107 4.64130 16544 28.9

19.540 4.53947 19454 34.0

19.972 4.44215 57202 100.0

20.794 4.26834 6267 11.0

21.211 4.18544 16460 28.8

21.949 4.04633 2025 3.5

22.761 3.90376 8352 14.6

23.546 3.77536 16771 29.3

24.244 3.66816 4461 7.8

24.545 3.62391 3873 6.8

24.944 3.56678 3373 5.9

25.539 3.48501 6776 11.8

26.192 3.39958 5690 9.9

26.566 3.35260 1841 3.2

27.451 3.24646 2866 5.0

28.132 3.16942 1300 2.3

28.458 3.13390 996 1.7 28.957 3.08101 1230 2.2

29.257 3.05010 2840 5.0

Based on the above data, the polymorph characterization of WO 01/34574 and Figure 21, it is understood that the polymorph of the dry mixture of Example 7 is mainly Form A but presence of minor quantity of Form B is also detected. In order to verify the possible presence of eriotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 13 times slower scan rate and the results are shown in Figure 22. The characteristic Form B peak was clearly observed. The relative intensity of this peak indicates that the percentage of Form B in API is approximately 0.4 %.

Example 8:

Eriotinib hydrochloride granules was analyzed by X-Ray powder diffraction. The X-Ray powder diffraction pattern along with the digitized patterns of polymorphs A and B of eriotinib hydrochloride Forms A and B of WO 01/34574, and the placebo pattern, can be found in Figure 23. The reflections are also set out in the following table.

14.330 6.17587 1097 2.8

14.726 6.01076 839 2.2

15.184 5.83047 653 1.7

15.619 5.66883 617 1.6

16.369 5.41091 6863 17.7

16.994 5.21317 1963 5.1

18.923 4.68584 9683 24.9

19.105 4.64172 15270 39.3

19.555 4.53588 22054 56.7

19.959 4.44497 38864 100.0

20.815 4.26404 7631 19.6

21.176 4.19214 10397 26.8

21.981 4.04050 2377 6.1

22.766 3.90295 9941 25.6

23.552 3.77436 18933 48.7

24.241 3.66869 5224 13.4

24.551 3.62309 4488 11.5

24.946 3.56655 3875 10.0

25.525 3.48693 6327 16.3

26.210 3.39735 4375 11.3

26.547 3.35501 2035 5.2

Based on the above data and the polymorph characterization of WO 01/34574, it is understood that the polymorph of the granules of Example 8 is mainly polymorph A but presence of minor quantity of Phase B was detected. In order to verify the possible presence of erlotinib hydrochloride Form B, the 2-theta region from 4-7 was scanned with a 13 times slower scan rate and the results are shown in Figure 24. From Figure 24 the characteristic erlotinib hydrochloride Form B peak is clearly observed. The relative intensity of this peak indicates that the percentage of phase B in API is approximately 0.9 %.

Example 9:

The following tablet compositions were prepared, in particular using pure eriotinib hydrochloride Form A as above as input API.

^: The actual quantity of eriotinib hydrochloride was calculated according to the assay and the water content of the drug substance lot being used. The quantity of lactose monohydrate was adjusted accordingly.

² Solvent used in granulation process; evaporated during the process and not present in the final product.

Example 10: The following PSD measurement was carried out.

Preparation of solutions:

Solution A: Accurately weigh 27.22g of monopotassium phosphate, dissolve in 1000 ml of water, and mix well.

Solution B: Accurately transfer 11.2 ml of saturated sodium hydroxide, dilute to 1000ml with water, and mix well

Buffer pH 7.4: Solution A - Solution B (50 : 39.1), mix well. Adjust pH to 7.4 with diluted phosphoric acid.

Sample solution: 200 mg of homogeneous erlotinib hydrochloride sample was mixed with Tween 80 (2 drops) and buffer pH 7.4 (4 drops) to obtain paste.

Procedure:

Refer to DQC-SOP 29A Use and Maintenance of Malvern Mastersizer 2000 and set the instrument parameters.

Add the sample solution to Hydro 2000 S(A) until the obscuration value lies within 10 - 20%.

Measure at 2500 rpm and 70% ultrasound (US) in Mastersizer dispersion cell after 4 minutes.