Introduction

Alkylating compounds, exemplified by cyclophosphamide monohydrate (CMH), represent potent chemotherapy agents widely employed in clinical settings. CMH is a staple treatment for various autoimmune disorders and malignancies, including ovarian, breast, lymphoma, and leukemia. Mechanistically, cyclophosphamide operates by impeding the replication of cancerous cells, thereby arresting their proliferation. This effect is mediated through the induction of DNA strand breakage followed by the formation of crosslinks within the cellular genetic material, ultimately culminating in cellular demise. Notably, the immunosuppressive properties of cyclophosphamide render it invaluable in organ transplantations and the management of select autoimmune conditions [1,2,3].

The commercially available products include cyclophosphamide for injection as a sterile dry powder and cyclophosphamide for injection (lyophilized with mannitol), offered in strengths of 500 mg, 1000 mg, and 2000 mg, necessitating reconstitution before patient administration. Cyclophosphamide can be administered intravenously as a bolus injection by reconstituting it with a 0.9% sodium chloride injection to achieve a concentration of 20 mg/mL. The sterile dry powder is needing vigorous agitation for extended time as part of reconstitution, but it maintains stringent control over its impurity profile, and which is tired process for healthcare professionals. In contrast, the lyophilized formulation with mannitol displays a increased impurity profile but offers a shorter reconstitution time [1,2,3,4,5].

Cyclophosphamide is available in two forms: monohydrate and anhydrous. The monohydrate form exhibits better physical and chemical stability. Commercially available injectable cyclophosphamide is supplied as a sterile dry powder, as a monohydrate form. To maintain the physical and chemical stability of cyclophosphamide monohydrate in its lyophilized condition, it is necessary to preserve a solid cake structure that does not shrink or melt [4, 5]. Throughout the lyophilization process, the drug product frequently assumes an amorphous form, rendering cyclophosphamide unstable. Therefore, to preserve the crystalline structure even post-lyophilization is challenging [6,7,8]. Following lyophilization, researchers explored non-traditional techniques to get the cyclophosphamide monohydrate. These methods entailed hydrating the drug product through the introduction of steam into the lyophilizer post-drying. or by bleeding it with nitrogen or humidified sterile air. These approaches pose some disadvantages, like uniformity and cake shrinkage or meltback [9, 10].

Cyclophosphamide is susceptible to degradation in aqueous conditions, resulting in its instability over time. As a result, it is impossible to create a readily usable water-based solution. Although several non-aqueous ready-to-dilute (RTD) formulations are available, they still necessitate dilution prior to administration, similar to the current practice with dry powder. Nonetheless, even in non-aqueous liquids, impurity profiles are elevated, and the utilization of significant quantities of organic solvents may present safety concerns for patients. Therefore, a lyophilized formulation of cyclophosphamide with reduced reconstitution time remains a viable alternative [2].

Existing study [8] have examined the retention of cyclophosphamide monohydrate during lyophilization in aqueous media at a drug concentration of 3.7%. To align with the marketed strengths, large fill volumes are necessary, equating to 13.5 mL for a 500 mg strength, 27 mL for a 1 g strength, and 54 mL for a 2 g strength. These substantial fill volumes result in prolonged lyophilization cycles. In response to these challenges, the proposed approach involved lyophilizing cyclophosphamide monohydrate utilizing non-aqueous solvents, either individually or in combination, to yield a drug product containing crystalline cyclophosphamide monohydrate post-lyophilization. This strategy aimed to attain a stable formulation and expedite the lyophilization cycle [11,12,13,14,15]. As a result, it led to decreased process time, improved product stability, and reduced costs, thereby ultimately benefiting patients by mitigating treatment expenses.

A literature review was conducted on the topic of lyophilization, focusing on the use of non-aqueous solvents. A patent application [11] was discovered that had certain restrictions in terms of an extended lyophilization cycle when using non-aqueous solvents, primarily in combination with water. Furthermore, there exists a particular formulation that employs acetone and tert-butyl alcohol (TBA) for the lyophilization procedure. This formulation comprises a solvent ratio of 1:0.75, with acetone being the primary constituent. Because acetone has a freezing point of -95 °C, it cannot freeze under typical lyophilizer operating conditions, which typically include a maximum freezing temperature of -60 °C and a maximum condenser temperature of -85 °C. Specialized technology, such as those using liquid nitrogen, can be used to freeze and solvent trap them in order to condense them during sublimation. However, this process is challenging and requires modifications to the lyophilizer machine, resulting in significant expenses in commercial manufacturing.

The primary goal of this research was to develop a stable lyophilized formulation of Cyclophosphamide monohydrate using non-aqueous solvents. Additionally, the selection of solvents based on the operational characteristics of the lyophilizer does not necessitate any modifications or additional configuration changes to the lyophilizer, resulting in reduced operational costs that were previously unexplored [16,17,18,19,20,21].

Materials and Methods

Materials

The donation of Cyclophosphamide monohydrate was graciously provided by MSN labs (India). Tertiary butanol, acetonitrile from Finar (Gujrat, India), lab-purified water was utilized, all additional chemicals used were of analytical grade.

Methods

Differential Scanning Calorimetry (DSC)

We employed a modulated differential scanning colorimetry, TA instruments Trios V5.7.2.101 to analyses separate bulk solutions containing different ratios of solvents, namely TBA and ACN (100:0, 95:05, 90:10, 0:100), each at a concentration of 200 mg/mL. Approximately 5 mg of samples were weighed in aluminum pans, which were then hermetically sealed [22,23,24,25]. A temperature ramp of 1 °C/min was utilized to gradually decrease the temperature from 25 °C to -60 °C, while recording thermal events. A heating rate of 1 °C/min with the modulation of ± 0.5 °C, every 30 s was applied, gradually raising the temperature from − 60 °C to 25 °C, while capturing thermal events to enhance the optimization of the lyophilization cycle. Data analysis was performed using Universal Analysis® software, version 4.5 A, by TA Instruments. As part of MDSC calibration, the equipment cell constant was performed by employing indium, while temperature and enthalpy calibrations were conducted using indium, tin, and water as reference standards.

Assay by HPLC

The potency was determined by high-performance liquid chromatography (HPLC) utilizing a Shimadzu system that has a UV absorbance detector set at 195 nm. The mobile phase consisted of a mixture of water and acetonitrile in a ratio of 70:30% v/v. The solute separation was achieved by utilizing a YMC AQ 250 × 4.6 mm, 5 μm column or equivalent, with a flow rate of 1.5 mL/min and a sample injection volume of 20 µL [5, 29]. The vial was opened by removing the seal and stopper. The solution was created by adding around 25 mL of diluent, and subsequent dilutions were made to a concentration of 500 ppm before being injected into the HPLC system and the method concluded after verifying the linearity (Supplementary Fig. 1), precision, and accuracy values as detailed in the Supplementary Table 1.

Related Substance by HPLC

Cyclophosphamide’s related substances were examined using an HPLC system with a UV absorbance detector set at 200 nm. The analysis utilized a reverse-phase column measuring 4.6 mm × 12.5 cm, packed with L76 material, operated at a temperature of 5 °C. The injection volume was set at 10 µL, and the flow rate was maintained at 1.2 mL/min. The mobile phase consisted of 0.2 mL of 85% phosphoric acid in 1 L of water, adjusted to a pH of 2.6. The diluent consisted of 7.5 mg/mL of mannitol in water. The area normalization method was employed for analysis, and known impurities (Cyclophosphamide-related compounds A, B and D) as well as unknown impurities were calculated [29].

Water Content by Karl Fischer (KF)

The water content was determined using a KF auto-titrator from Metrohm. The titration flask was filled with an appropriate amount of methanol and then titrated with Karl Fischer reagent to neutralize the methanol [26,27,28,29]. A syringe was used to transfer 10 mL of neutralized methanol from the KF titration flask into the product vial. The contents were dissolved and subsequently extracted using the same syringe. The resulting solution was then transferred into the KF vessel for titration in order to determine the end point. The moisture content in W/W was subsequently calculated using the formula below.

$$\:Water\:content=\frac{Titre\:value\:in\:mL\times\:KF\:factor}{534.5 mg}\:\times\:100$$
Reconstitution time

To reconstitute the lyophilized product, a vial was taken, and the flip-off seal was removed. Then, 25 mL of water for injection or 0.9% sodium chloride was added to the vial [5, 8]. The vial was shaken to completely dissolve the contents, and the time was recorded to dissolve the complete solid content.

Residual Solvent by GC

Residual solvents were analyzed using a gas chromatography SHIMADZU / GC2030 model. It utilized a 6% Cyanopropylphenyl-94% dimethylpolysiloxane column (DB-624) measuring 30 m in length with an inner diameter of 0.53 mm and a film thickness of 3 μm, or an equivalent column. Helium served as the carrier gas at a flow rate of 4.0 mL/min, controlled in linear velocity mode. The detector temperature was set at 300 °C, with a split ratio of 1:5. The run time for each analysis was 20.0 min. The diluent solution used was DMSO [11].

Bulk Solution Hold Studies

A cyclophosphamide monohydrate solution with a concentration of 200 mg/mL was prepared by using a solvent system consisting of 90% TBA and 10% ACN. Additionally, a solution of cyclophosphamide monohydrate at a concentration of 37 mg/mL was produced using water. These bulk solutions were stored at 25 °C and 2–8 °C for 24 h before undergoing analysis for physicochemical properties, including description, assay, and impurity profile [30].

Filtration, Filling, Stoppering & Loading

The bulk solution batch size of 150mL was manufactured with a solvent ratio of TBA to ACN (90:10) at a concentration of 200 mg/mL. It was then filtered using PVDF filters and filled into 50mL USP Type I tubular vials with a 2.5mL fill. About 30-40nos. of vials were half stoppered with bromo butyl slotted rubber stoppers and placed in a Lyophilizer (SP Scientific) for lyophilization on pre-cooled shelves at 5 °C [8, 11].

Lyophilization

The lyophilization conditions were established by evaluating the bulk solution by DSC and conducting vapor pressure calculations by employing the Clausius-Clapeyron Eqs. [31,32,33,34,35,36,37]. The loaded vials were lyophilized at various freezing conditions and explored to determine optimal parameters, including freezing with and without annealing. Due to the inclusion of acetonitrile (ACN) in the solvent system, the freezing temperature was set to -55 °C, which is 10 °C below the freezing point of ACN (-45 °C). Freezing was conducted at ramp rates of 1 °C/min and 0.5 °C/min, with annealing, to evaluate the impact of the freezing rate. Following freezing, the primary drying was evaluated at temperatures ranging from − 55 °C to -10 °C under varying vacuums. Mass transfer is studied from − 55 °C to -10 °C by unloading the samples at predetermined time points using sampling thief, and the samples were examined for gravimetric and physical appearance. We measured the amount of water and residual solvent in the optimized lyophilization cycle, as well as its physical and gravimetric appearance. This helped us figure out when the cycle ended and set the final parameters for the lyophilization cycle [38,39,40,41].

Stability Study

The optimized lyophilization cycle formulation underwent both accelerated and real-time stability tests. These tests involved storing the formulation at 25 °C/60% RH and 2–8 °C for a duration of 6 months [22, 30]. Samples were collected at designated intervals (1, 2, 3, and 6 months) and subjected to analysis for Appearance, Reconstitution time, Water content, potency, and related substances. The stability study was to identify any significant deviations from the initial values.

Results and Discussion

Solubility Study

The solvent chosen for this process needed to meet specific criteria, including excellent solubility, manufacturing feasibility, and chemical stability. The solubility of CMH in various solvents (aqueous and non-aqueous) and the dissolution behaviour captured in Supplementary Table 2. The findings show that CMH is substantially more soluble in non-aqueous solvents than in water. The water solubility of CMH was found to be around 40–50 mg/mL, which is in line with the values documented in previous studies. In contrast, the solubility of CMH in acetonitrile and ethanol was found to be approximately 500 mg/mL, which is higher than the values reported in the existing literature [11]. This substantial difference illustrates the preference of CMH for dissolving in non-aqueous solvents. Moreover, the sublimation of non-aqueous solvents is faster due to their high vapor pressure in comparison to water.

Bulk Solution Optimization

The manufacturing feasibility was studied based on the freezing point of solvents, which was a crucial factor in solvent selection. The literature provides extensive documentation on the freezing points of various solvents. In particular, water freezes at 0 °C, tert-butyl alcohol (TBA) at room temperature, acetonitrile (ACN) at -45 °C, and ethanol at -114 °C captured in Supplementary Table 2 [42]. Despite ethanol’s good solubility, the lyophilizer’s maximum operating temperature of -60 °C, insufficient for freezing, discourages its use.

Tert-butyl alcohol (TBA) is commonly used as a co-solvent in lyophilization procedures. Nevertheless, the use of this solvent alone has limitations due to its tendency to solidify at ambient temperature, creating difficulty in manufacturing bulk solutions at a temperature range of 2 °C to 8 °C. Higher temperatures increase the risk of impurities, as thermal degradation is one of the main modes of cyclophosphamide degradation [5, 43,44,45].

Neat ACN has a freezing point near − 45 °C. When drug is dissolved in pure ACN, the DSC analysis showed an exothermic peak at -53.44 °C (Supplementary Fig. 2), which means a freezing point depression of about 10 °C when the drug is dissolved in ACN, which is equivalent to the maximum temperature at which lyophilizers operate. Therefore, ACN alone is not suitable as a solvent for the lyophilization process.

An alternative approach to improving manufacturability is to use a combination of TBA and ACN while considering their individual freezing points. The freezing point depression of TBA has been determined by combining TBA with ACN solvents in different ratios. Freezing was observed in the ratio of 95:05 when kept at temperatures of 2 to 8 °C. However, no freezing occurred in the TBA: ACN ratios of 80:20, 85:15, or 90:10, and these observations are shown in the Supplementary Fig. 3. Because the TBA: ACN 95:05 solvent ratio has a tendency to freeze at 2 and 8 °C, bulk solution manufacturability is low. Additionally, given the freezing point of ACN, it is recommended to use smaller proportions in the bulk, which means that the 80:20 and 85:15 ratios should be eliminated. Therefore, the main emphasis on optimizing the bulk solution switched to the ratio of 90:10 TBA: ACN [42].

Furthermore, DSC findings (Supplementary Fig. 4) of TBA: ACN mixture with a 90:10 ratio indicate there is an exothermic peak at 0.2 °C, which is in line with the observation of no freezing when it was stored at 2–8 °C (Supplementary Fig. 3). Hence, this ratio demonstrated excellent manufacturing feasibility, leading to the selection of the TBA: ACN 90:10 ratio as the final bulk solvent composition. DSC cooling curve analysis of Cyclophosphamide bulk solution at 200 mg/mL in a TBA: ACN 90:10 solvent ratio revealed an exothermic peak at -16.14 °C (Supplementary Fig. 5), which indicates the minimum temperature needed for complete freezing. Additionally, it possesses a high vapor pressure, leading to an accelerated sublimation rate at lower temperatures in the primary drying phase. In addition, TBA and ACN demonstrate negligible toxicity [46].

Bulk Hold Studies

Bulk hold time data, meticulously documented in Table I, provided critical insights into the stability of Cyclophosphamide formulations under varying conditions. The findings from studies on bulk hold unveiled an unexpected result: solutions prepared with pure water exhibited accelerated degradation rates. Notably, Cyclophosphamide-related compound B surpassed the stringent USP limit of 0.15% at both 2–8 °C and 25 °C after a mere 24-hour period.

Table I Cyclophosphamide injection bulk solution hold time data
Full size table

In stark contrast, bulk solutions prepared with pure, non-aqueous solvents exhibited a markedly superior impurity profile. The levels of Cyclophosphamide-related compound B remained impressively low, measuring at a mere 0.06% at 2–8 °C and 0.09% at 25 °C after the same 24-hour duration, well below the stringent USP limit of 0.15%.

These findings underscored the critical importance of selecting the appropriate solvent system to ensure the stability and integrity of Cyclophosphamide formulations during bulk storage. In light of these results, it became evident that a non-aqueous bulk solution maintained at 2–8 °C emerged as the optimal approach for effectively controlling Cyclophosphamide degradation at the bulk stage, safeguarding the quality and efficacy of the final pharmaceutical product.

Lyophilization

The lyophilization process, by retaining the cyclophosphamide monohydrate, must preserve the bound water within the final product to ensure its stability and effectiveness. To achieve this, freezing and drying conditions were carefully optimized based on the thermal events of DSC.

Optimization of Freezing

The freezing temperature was determined using DSC cooling curve observation, which revealed an exothermic peak at -16.14 °C (Supplementary Fig. 5) attributed to ice crystallization. Since ACN is a component of the bulk solution, there is a possibility of ACN phase separation during the freeze concentrate process. Hence, to ensure the product’s integrity and stability, it is advised to maintain a minimum temperature of -53 °C during the freezing process, as the drug in pure ACN exhibits an exothermic peak at -53 °C (Supplementary Fig. 2). The freezing stage of the lyophilization process was conducted at a temperature of -55 °C, with a gradual rise of 1 °C per minute over a duration of 60 min, followed by a 120-minute period of maintaining the temperature. The durations were carefully selected to guarantee the system’s complete freezing [8, 11].

Optimization of Primary Drying

After the freezing phase, the drying process began with a 150 mTorr vacuum and a temperature of -55 °C. The selected vacuum for sublimation was well below the minimum vapor pressure needed for this solvent system at a particular temperature, as determined by vapor pressure calculations in Supplementary Tabel III. The initial temperature of -55 °C ensures the maximum removal of the ACN and reduces the risk of a product microcollapse. This is based on the endothermic peak at -47.18 °C (Supplementary Fig. 6) that is observed in the DSC heating curve of the drug when it is dissolved in pure ACN. This step involved a 60-minute ramp to reach the desired vacuum, followed by a 360-minute hold period and further drying temperature were selected considering the DSC heating curve (Supplementary Fig. 7) of optimized bulk solution.

After the initial drying step at -55 °C/150 mTorr, the temperature and vacuum were increased to -30 °C/300 mTorr in 360 min and held for 3000 min. The gravimetric analysis findings from the mass transfer study samples indicate that the quantity of solvent that underwent sublimation led to an extended cycle. Despite an 1800-minute drying period at -30 °C, total mass transfer was not attained, and full drying occurred only after 2400 min (Table IIA). Therefore, it is not feasible to use a drying temperature of -30 °C in order to achieve a quicker lyophilization process.

Table II Mass transfer data (n = 3 vials)during lyophilization process
Full size table

In order to expedite the lyophilization process, the temperature and vacuum were adjusted to -25 °C/300 mTorr over a period of 360 min and maintained for 1440 min [47, 38]. During this stage, samples for mass transfer were taken at specific time intervals. The data (Table IIB) clearly demonstrates that a temperature change significantly increased the rate of sublimation. After a duration of 720 min, the samples exhibited complete drying. Furthermore, even after extending the duration to 1440 min, no bound moisture was eliminated.

In order to assess the impact of temperature on sublimation, the temperature was further raised to -10 °C/300 mTorr over a period of 120 min. Samples were collected at 60 min and 120 min. The results of the physical description show that the cake structure shrinks; this might be due to the low dehydration enthalpy of CMA, and the water content measurements of 2.33% and 4.76% further substantiate the removal of the bound water (Figs. 1 and 2). These findings indicate that it is not advisable to use higher drying temperatures [40].

Fig. 1
figure 1

In process water content of vials at various stages lyophilization process

Full size image
Fig. 2
figure 2

Finished product images at various stages, a Sample liquified b Finished product with good cake, c Lyophilized vials with partially melted cake

Full size image

Lyophilized Product Testing

The final product data meets the criteria for appearance, showing a well-formed cake with a water content of about 6.5% and solvent levels of TBA and ACN that are in line with ICH guidelines. This is true even though the sublimation process at -25 °C shortens the time it takes to freeze. Nevertheless, the vials, as indicated in Table III, demonstrated a reconstitution time beyond 2.5 min, which is going to be a tiring process for healthcare professionals. This finding necessitated investigation, and additional enhancements to the cycle are required to reduce the reconstitution time.

Table III Analytical data of product sublimated at -25°C in primary drying without annealing step in freezing
Full size table

Investigation on Higher Reconstitution Time

Based on existing research, by incorporating the crystallization temperature into the freezing process as a form of annealing, it will facilitate the growth of large crystals. This, in turn, will promote the development of larger pore sizes and improve the efficiency of the sublimation rate, leading to better cake with a quick reconstitution time. Moreover, annealing can enhance the crystallization of cyclophosphamide monohydrate during freezing, help preserve monohydrate, and also enhance porosity. To incorporate annealing approach DSC heating curve indicating an exothermic peak at -22.16 °C (Supplementary Fig. 7) which revealed an attributed crystallization peak. Hence, the freezing cycle was optimized by introducing the crystallization temperature as part of freezing with finalized drying parameters. Therefore, the operation of lyophilization was carried out by setting a crystallization temperature of -22 °C, as specified in the cycle parameters provided in Table IV. Following the completion of the cycle (Fig. 3), samples were obtained from different locations and examined for physicochemical characteristics. The information shown in Supplementary Table 4 demonstrates that the issue of prolonged reconstitution time was successfully addressed without impacting other physical chemical characteristics. The enhanced lyophilization procedure has effectively reduced the time required for reconstitution to under 30 s by generating a porous cake structure during the annealing stage. Furthermore, the final results of the finished product mentioned in Supplementary Table 4 adhere to product specifications (Supplementary file Table 5) and regulatory standards in ensuring the final product’s safety and quality [8, 48,49,50].

Table IV Optimized Lyophilization Cycle parameters
Full size table
Fig. 3
figure 3

Lyophilization cycle history data

Full size image

XRD Analysis Data

The X-ray powder diffraction (XRPD) patterns of the received cyclophosphamide monohydrate (CMH) and the lyophilized formulations were analyzed. The XRPD pattern (Fig. 4a, b) shows characteristic peaks at 2θ values of 7.0, 14.1, 14.7, 17.8, 23.6, 26.7, and 28.2, which correspond to the API. This confirms the presence of CMH in the lyophilized sample, maintaining a good cake structure. Additionally, the 2θ values of CMH align with existing literature [8, 51, 52]. Samples exhibiting partial melting showed reduced intensity for these characteristic peaks, while completely melted samples did not display any 2θ values, indicating a full transformation to an amorphous state.

Fig. 4
figure 4

XRD data, Fig. 4a (A As such API, B In-process sample, C Finished product initial & D Finished product stability, E Drug product partially melted and F complete melted/ collapsed cake, Fig. 4b zoomed version of API

Full size image

Stability Study

The optimized lyophilization cycle formulation was studied for accelerated and long-term stability testing as per ICH for a period of 6 months. The tests were performed at a temperature of 25 °C and a relative humidity of 60%, as well as in a temperature range of 2–8 °C. This investigation entailed a comprehensive assessment of multiple factors, including the visual appearance of the cake, reconstitution time, water content, potency, and the related substances. According to the data in Table V, these factors did not undergo any significant changes. XRD analysis of the stability sample also confirms that the 2 θ values are the same as the initial samples. This means that the monohydrate form has been retained. This suggests that we can consider the formulation stable.

Table V Stability data of optimized lyophilization cycle formulation
Full size table

Conclusion

After a thorough investigation, the optimized bulk solution with non-aqueous solvents TBA to ACN at a 90:10 ratio at a concentration of 200 mg/mL improves solubility, stability, and manufacturability. Furthermore, it significantly reduces the fill volume and shortens the lyophilization process. The lyophilization cycle highlights how freezing with annealing significantly reduces the time required for reconstitution. Mass transfer investigations emphasize the significance of selecting the appropriate drying temperature. A temperature of -30 °C at 300 mT prolongs the drying process, whereas an increase in temperature to -10 °C at 300 mT leads to dehydration, resulting in shrinkage and collapse. In addition, maintaining a drying temperature of -25 °C at 300 mT allows for a shorter drying process and preserves the monohydrate form. The results confirm the retention of the cyclophosphamide monohydrate form at initial and on stability, while also meeting all other specification parameters. This demonstrated the success of formulation development using a reduced lyophilization cycle.