Amorphous solid dispersions of praziquantel: can the active R-enantiomer alone outperform the racemic tablet?

Background and motivation

Praziquantel (PZQ) is the drug the World Health Organization recognizes as the first choice against schistosomiasis, a parasitic disease caused by Schistosoma flatworms that has affected more than 250 million people (Section 1, refs 1–4) C1. PZQ is a Biopharmaceutics Classification System (BCS) class II drug: it permeates membranes well but dissolves poorly, with a cited water solubility of 0.4 mg/mL at 25 °C, so its oral bioavailability is limited by slow dissolution in the gut (refs 5–9) C2·M2. Together with extensive first-pass metabolism, this is why the usual dose is high, 20–40 mg/kg C2.

The marketed tablets contain the crystalline racemic compound (RS-PZQ), a crystal in which the R- and S-enantiomers sit 1:1 in one lattice. The therapeutic effect comes mainly from R-PZQ; the S-enantiomer contributes an unpleasant bitter taste and doubles the size of the tablet, both of which hurt compliance (Meyer 2009, ref 3; Olliaro 2014, ref 10; Kovač 2017, ref 11) C1. A formulation of R-PZQ alone that dissolves well would therefore address two problems at once: poor dissolution and the burden of the inactive enantiomer. An affordable R-PZQ formulation has indeed been named a priority (ref 3) C38.

Many routes have been tried to make PZQ dissolve faster: cocrystals with carboxylic acids or with niclosamide, hydrates and solvates, new anhydrous polymorphs, cyclodextrin complexes, micro- and nanocrystals, nanosuspensions, solid lipid and polymeric nanoparticles, clay and calcium carbonate carriers, liposomes and implants (refs 12–31). Solid dispersions of PZQ with sodium starch glycolate, sodium alginate, carboxymethylcellulose, polyvinylpyrrolidone, PVPVA, polyethylene glycol and poloxamer have improved its biopharmaceutical properties (refs 9, 45–48) C4. What those studies did not do, "to the best of our knowledge" in the authors' words, is examine how PZQ is released from amorphous solid dispersions (ASDs), and none had tested an ASD containing only R-PZQ C4.

ASDs, in which the drug is molecularly dispersed in a polymer in the amorphous state, are widely used to raise solubility, dissolution rate and the extent of supersaturation, and through these permeability and bioavailability (refs 32–44) C3. The weak point of the approach is crystallization: a supersaturated solution produced by an ASD can crystallize back to the stable crystal. Polymers dissolved in the medium can delay this by interacting with the drug and suppressing nucleation and/or crystal growth (Sarode 2014, ref 51; Ueda 2014, ref 52), and cellulose derivatives such as HPMCAS and HPMC are well established inhibitors across many drugs (refs 52, 56–63) C5·C24. A further twist is that cellulose polymers are themselves chiral and have been shown to inhibit crystallization of the two enantiomers of a drug to different extents in amorphous dispersions (Miyazaki 2011, ref 65; Sato and Taylor 2015, ref 66) C28.

The paper sets out to formulate stable PZQ ASDs that release the drug better than the commercial tablets, to learn how polymer type, preparation method, drug loading (DL), drug concentration, medium pH and chirality affect release, and to find out whether ASDs of R-PZQ alone release the drug as efficiently as racemic ASDs or more so (Section 1) A1–A8.

Background concepts

Crystalline and amorphous solubility. The crystalline solubility is the concentration in equilibrium with the crystal; it is the lowest concentration a solid form can maintain over time. The amorphous solubility is the highest concentration of dissolved drug that the amorphous form can reach; adding more drug beyond it makes a separate drug-rich liquid or amorphous phase appear, which scatters light. The authors measure it by the UV-extinction method of Ilevbare and Taylor (ref 49): a concentrated DMSO solution of the drug is pumped slowly into buffer, and the concentration at which scattering at a non-absorbing wavelength (400 nm) rises sharply is taken as the amorphous solubility P2. The ratio of the two solubilities is the largest supersaturation an amorphous formulation can produce. For PZQ in pH 6.5 buffer at 37 °C, the crystal values are 282.6 µg/mL for R-PZQ and 196.4 µg/mL for RS-PZQ, and the amorphous values 1.4 and 1.1 mg/mL, so the amorphous form offers roughly a fivefold gain R1·R2·R17.

Nucleation induction time. A supersaturated solution stays clear for some time before crystals appear. That time, the induction time, measures how long the solution resists crystallization; in this paper it is detected as the rise in UV extinction at 400 nm as crystals nucleate and grow P3. Comparing induction times with different dissolved polymers ranks the polymers as crystallization inhibitors. Because the measurement here is capped by the length of the run (83 min), two polymers that both reach the cap cannot be ranked against each other R3.

Racemic compound, enantiomers and crystal forms. RS-PZQ crystallizes as a racemic compound; its most stable, commercial form is an anhydrate in the triclinic space group printed as P1, in which all molecules adopt the syn conformation of the carbonyl groups (ref 12) C26·M2. R-PZQ on its own is known only as a hemihydrate and a monohydrate, both monoclinic C2, with the anti conformation (refs 3, 64) C26·M1. R-PZQ also melts lower, at about 110 °C, against about 142 °C for the racemic crystal (refs 50, 70, 71) M1·M2. Different crystal forms mean that the crystal that limits solubility, and the crystal that nucleates from supersaturated solution, can differ between the racemate and the enantiomer.

Release patterns and the "cliff". In ASD dissolution tests the fraction of drug released often depends on DL. In some systems release collapses above a threshold loading, which the field calls "falling off the cliff" (Indulkar 2019, ref 67); in others it decreases gradually with DL, a pattern earlier work linked to release controlled by the polymer (Saboo 2020, ref 68) C16·C17. "Polymer-controlled" means the drug leaves at the rate at which the polymer dissolves.

Nonsink tests and target concentration. The release tests here are nonsink: the amount of ASD added would give a target concentration (667 µg/mL or 1.5 mg/mL) well above the crystalline solubility, so the test asks how much supersaturation the formulation can generate and hold P8. The higher target, 1.5 mg/mL, is close to the R-PZQ amorphous solubility (1.4 mg/mL) but above that of RS-PZQ (1.1 mg/mL) C17·R2.

Enteric and acid-soluble polymers. The authors describe HPMCAS-MF as an enteric polymer of rather hydrophobic nature with limited solubility below pH 6, so it stays largely undissolved in gastric fluid (refs 53–55) M6. HPMC, by contrast, is readily soluble in acidic media that simulate the stomach (ref 76) M8. A two-stage test (an acid stage followed by a shift to intestinal pH) shows how this difference plays out for release P10.

Methods in detail

The study has three experimental parts that build on each other: measure the solubilities that set the supersaturation scale, screen polymers by how long they keep supersaturated PZQ from crystallizing, and then make ASDs with the two best polymers and test their release against the commercial tablets under a series of conditions. All tests ran at 37 °C.

Solubility measurements

For the crystalline solubility, excess crystalline R-PZQ or RS-PZQ (described as "in hemihydrate form") was stirred at 300 rpm and 37 °C for 48 h in PBS pH 6.5; excess solid was removed by ultracentrifugation (35,000 rpm, 37 °C, 30 min) and the supernatant assayed by HPLC with UV detection at 220 nm against a 1–100 µg/mL standard curve P1. The number of replicates is not stated P1. For the amorphous solubility, 125 mg/mL DMSO stock solutions were added at 50–200 µL/min to 50 mL of PBS pH 6.5 or HCl solution pH 1.6 in water-jacketed beakers stirred at 400 rpm, while scattering at 400 nm was recorded every 5 s with a 2 mm probe P2. The PBS was made from NaOH, sodium phosphate monobasic monohydrate and NaCl (4 g, 31.6 g and 49.42 g per 4 L); the HCl solution from 8 g NaCl and 5.6 mL concentrated HCl per 4 L M14·M15.

Polymer screen by induction time

The problem this step addresses is choosing a carrier that will keep a supersaturated PZQ solution from crystallizing. Seven polymers were each predissolved at 300 µg/mL in PBS pH 6.5: HPMCAS-LF and HPMCAS-MF (AQOAT grades), HPMC PHARMACOAT 603 and HPMC Methocel E5 Premium LV, hydroxypropyl cellulose (HPC, KLUCEL EXF), PVPVA (Kollidon VA 64) and poly(vinyl alcohol) (PVA, Parteck MXP) (Figure 1) M5–M11·P3. Drug in DMSO was then added up to the amorphous solubility of the drug being tested, 1.1 mg/mL for RS-PZQ and 1.4 mg/mL for R-PZQ, and extinction at 400 nm was followed every 5–30 s at 37 °C and 400 rpm P3·R2. RS-PZQ was run in triplicate with all seven polymers and without polymer; R-PZQ was run in duplicate with the two best polymers and without polymer (Figure 2) P3.

Note by the AI writer (not discussed in the paper): because each drug was tested at its own amorphous solubility, the two drugs were at different absolute concentrations. Relative to their own crystalline solubilities, the supersaturation ratios are about 5.0 for R-PZQ and about 5.6 for RS-PZQ, so R-PZQ was, if anything, at slightly lower supersaturation X14.

ASD preparation and solid-state checks

Solvent-evaporated (SE) ASDs were made by dissolving drug and polymer in dichloromethane/methanol 1:1 v/v, removing the solvent in a rotary evaporator with the water bath at 50 °C, drying overnight under vacuum at ambient temperature, cryomilling and sieving through a 60 mesh (250 µm) screen P4. RS-PZQ was formulated at 20, 30, 40 and 50% DL and R-PZQ at 30 and 40% DL, with HPMCAS-MF and with HPMC E5 LV P4. Hot-melt extruded (HME) ASDs of RS-PZQ with HPMCAS-MF (20 and 30% DL) were made in an Xplore PME extruder with a 5 mL barrel and conveying screws at 145 ± 1 °C, 50 rpm and 5 min residence time, then cryomilled and sieved in the same way P5. The authors chose to heat above both the melting point of RS-PZQ (about 142 °C) and the glass transition temperature of HPMCAS-MF to obtain the highest degree of molecular mixing C29·C30.

Every ASD was checked by X-ray powder diffraction (XRPD, Cu Kα, 4–40° 2θ, 0.02° step) P6. For physical stability, selected ASDs were stored in loosely capped vials in a sealed container held at 40 °C and 75% relative humidity (RH) by saturated NaCl; the patterns shown are for the 20% DL SE RS-PZQ ASD with each polymer, after 21 weeks (HPMCAS-MF) and 23 weeks (HPMC E5 LV) P7.

Dissolution tests

Powder dissolution was run in triplicate in a USP II apparatus at 37 °C and 100 rpm, with the drug concentration followed in situ by fiber-optic UV probes (second-derivative baseline correction, calibration on the area under the 278–285 nm band, 10–1500 µg/mL) P8. In single-stage tests, ASD powder was added to 50 mL PBS pH 6.5 at a target of 667 µg/mL or 1.5 mg/mL and followed for about 2 h P8. A 600 mg tablet was tested in 400 mL PBS pH 6.5, giving the same 1.5 mg/mL target P9. In two-stage tests, ASDs were first held in 45 mL HCl solution pH 1.6 for 1 h, then 5 mL concentrated PBS pH 7.3 was added to bring the pH to 6.5; tablets were tested in 360 mL acid followed by 40 mL PBS pH 7.3 P10·P11. The ASDs in the two-stage test were SE RS-PZQ ASDs with HPMCAS-MF at 20 and 30% DL and, according to the Figure 5 legend only, with HPMC E5 LV at 40% DL P10·X9.

Note by the AI writer (not discussed in the paper): the tablets and the ASDs were not tested in the same set-up. A whole tablet in 400 mL and sieved powder below 250 µm in 50 mL differ in dosage form, surface area and hydrodynamics, even at the same target concentration X5.

Experimental analysis

The two cellulose polymers hold off racemate crystallization longest

Without polymer, RS-PZQ at its amorphous solubility crystallized after 5.5 ± 1.3 min (Figure 2a) R3. Every polymer lengthened this time, in the order PVPVA, HPC, PVA, HPMC 603, HPMCAS-LF, and HPMCAS-MF and HPMC E5 LV prevented crystallization for the whole 83 min run (Table A) R3·C7. The authors chose these two as ASD carriers C7. They attribute the shorter times of the other grades of the same polymers, HPMCAS-LF and HPMC 603, to differences in "functional group content and amphiphilicity" that "likely" change how the polymers interact with nascent crystals; neither property was measured C25.

Table A. Nucleation induction times in PBS pH 6.5 at 37 °C, each drug at its own amorphous solubility (from Section 3.3 and Figure 2; the R-PZQ value without polymer is read from Figure 2b, approximate) R3·R4.

Both polymers protect R-PZQ much less well

When the two best polymers were tested with R-PZQ, the picture changed (Figure 2b). HPMCAS-MF held R-PZQ for 37 ± 11 min and HPMC E5 LV for 15.5 ± 1.6 min, against at least 83 min for the racemate with either polymer R4·R3. Without polymer the two forms crystallized equally fast, at about 5 min R4.a·C9. The authors describe the inhibition of R-PZQ as "significantly less potent" for HPMCAS-MF and "significantly less" for HPMC E5 LV C8. No statistical test is reported, and the R-PZQ values rest on two runs each X12. The size of the drop, however, is large: from beyond the 83 min cap to 37 and 15.5 min.

The ASDs are amorphous when made; stability on storage differs

All freshly made ASDs, SE and HME, racemic and enantiopure, gave only broad halos in XRPD, while the starting crystals of R-PZQ and RS-PZQ gave sharp, different patterns (Figure 3a–g) R5·R6. On storage at 40 °C/75% RH, the 20% DL HPMC E5 LV ASD stayed halo-only after 23 weeks, while the 20% DL HPMCAS-MF ASD "exhibited small crystalline peaks within 21 weeks" (Figure 3h, i) R7·C11. In Figure 3h these appear as small sharp features near 15, 16.5–17 and 20° 2θ on the amorphous halo R7.

Note by the AI writer (not discussed in the paper): the text treats these results inconsistently. Section 3.4 states that all patterns in Figure 3e–i show only halos, the next sentence reports crystalline peaks in Figure 3h, and the abstract, Section 4.1 and the Conclusions say that the ASDs of both polymers "remained physically stable for several months" C10·C11·C12·X1. The storage test also covered only one 20% DL ASD per polymer, with no R-PZQ, HME or higher-DL ASD X2.

Release at 667 µg/mL: complete up to high loading

At the lower target, 667 µg/mL, which lies below the amorphous solubility of both forms, almost every SE RS-PZQ ASD released all its drug (Figure 4a) R8·R2. HPMCAS-MF ASDs released completely at every DL from 20 to 50%, within 10–21 min at 20 and 30% DL R8·C13. HPMC E5 LV ASDs released completely up to 40% DL, more slowly (28–53 min), and at 50% DL levelled off at about 86% R8·C13. Read from the figure, several "complete" plateaus lie above 100%: about 112% for HPMCAS-MF at 20% DL and about 108% at 30% DL R8.a–b·X3. The authors read these results as showing that both polymers are carriers "capable of facilitating the effective release of the drug" and see no "falling-of-the-cliff" effect, since release "never declined massively" with DL C15·C16.

Release at 1.5 mg/mL: a gradual decline with HPMCAS-MF

At 1.5 mg/mL, close to the R-PZQ amorphous solubility, the DL effect appears (Figure 4b, Table B) R9·C17. HPMCAS-MF RS-PZQ ASDs released completely within 51 min at 20% DL and levelled off at 94% and 83% at 30 and 40% DL R9.a–c. The authors call this decline gradual, note that it matches earlier observations (Saboo 2020, ref 68), and say it "suggests that the release is polymer controlled" C17. Polymer release was not measured, so this reading rests on the pattern, not on a direct comparison of drug and polymer release C17.

Figure 4b also contains two HPMC E5 LV series that the text does not describe. The 30% DL ASD stayed at about 74–77%, while the 20% DL ASD reached about 69% at 10 min and drifted down to about 55.5% by 110 min, with error bars reaching down to about 24% R9.d–e·X11. Note by the AI writer (not discussed in the paper): this lower release at the lower loading, with large scatter, is not what "robust release" from both polymers would suggest at this concentration X11.

Table B. Drug release (%) in 50 mL PBS pH 6.5 at a target of 1.5 mg/mL, read from Figures 4b, 4d, 4e and 4f (approximate; mean of n = 3). The text gives 94% and 83% for the HPMCAS-MF RS-PZQ ASDs at 30 and 40% DL, 94% and 77% as the maximum for the R-PZQ ASDs at 30% DL, and 62% as the final value of the R-PZQ HPMC E5 LV ASD R9·R11·R12·R13.

Note by the AI writer (not discussed in the paper): a release of 100% of a 1.5 mg/mL target means a concentration of about 1.5 mg/mL, above the measured amorphous solubility of RS-PZQ in PBS (1.1 ± 0.1 mg/mL). Any RS-PZQ value above about 73% therefore exceeds that solubility. The in situ UV signal may then include drug in a colloidal or polymer-associated form, a point the paper does not discuss X4·R2.c.

Chirality: R-PZQ keeps up with HPMCAS-MF but not with HPMC E5 LV

With HPMCAS-MF at 30% DL, the R-PZQ and RS-PZQ ASDs gave overlapping release profiles that both reached 94% (Figures 4e, 4f) R12.a·R9.b·C39. At 40% DL, R-PZQ reached about 92.5% against about 82.5% for RS-PZQ at 110 min (Figure 4f) R13.a·R9.c. The authors write that "an enhanced release of R-PZQ from HPMCAS-MF ASDs was observed" at 40% DL and call the 30% DL result "promising" C39. For R-PZQ, raising DL from 30 to 40% slowed the early release (about 76% against 67% at 10 min) without lowering the plateau R12.a·R13.a·C18. Note by the AI writer: the RS-PZQ error bar at 40% DL is about ±11 percentage points at 110 min, roughly the size of the gap, and no statistical test is given; the R-PZQ advantage is best read as a trend X13.

With HPMC E5 LV at 30% DL, R-PZQ behaved differently from the racemate. Its release peaked at 77% at about 25 min and then fell to 62%, while the RS-PZQ ASD with the same polymer stayed at about 74–77% (Figure 4e) R12.b·R9.e. The authors attribute the fall "likely" to crystallization and note in Section 4.2.4 that it "hinted toward" crystallization, in line with the short R-PZQ induction time with HPMC E5 LV (15.5 min) C20·R4.b. No crystals were looked for in the dissolution vessel, so this remains an inference from the release curve C20. From this comparison the authors conclude that HPMCAS-MF "was identified as the preferred polymer system" for R-PZQ C40.

Hot-melt extrusion

The HME ASDs of RS-PZQ with HPMCAS-MF released completely at both DLs and both targets, in times comparable to the SE ASDs (Figures 4c, 4d) R10·R11·C31. At 1.5 mg/mL they reached about 106% (20% DL) and 100% (30% DL), against about 100% and 94% for the SE ASDs; at 667 µg/mL the plateaus were about 117% (HME) against 112% (SE) at 20% DL and about 108% for both at 30% DL R11·R9.a–b·R10·R8.a–b. The authors call the HME advantage "slightly better release" and see HME as a promising manufacturing route; R-PZQ, which melts at about 110 °C, was not extruded C31·C32. Note by the AI writer: these differences are of the same size as the excess above 100% in several "complete" plateaus, so the ranking of HME above SE is not resolved by these data X3.

Commercial tablets

The generic USP tablet and the branded Biltricide tablet each released about 14–16% by 10 min and stayed there to 110 min; the text gives a maximum of 16% (Figure 4g) R14. Against the ASDs, which reached more than 90% at 1.5 mg/mL with DLs up to 40% (R-PZQ), the authors put the maximum gain "in the order of 6.25 times" (abstract: "of the order of 6 times") and say this is "in line with the amorphous solubility advantage, which is around fivefold" R16·R17·C42·C43. The fivefold figure matches R-PZQ in PBS (1.4 mg/mL against 282.6 µg/mL); the paper does not say which pair of values it used R17·R1·R2.a. The 6.25-fold figure corresponds to complete release against 16%; for the 40% DL R-PZQ ASD (about 92.5%) the ratio is about 5.8 X5.

The pH-shift test

In the two-stage test, the polymer decided when release happened (Figure 5, Table C) R15. In 45 mL HCl pH 1.6, the HPMCAS-MF ASDs released slowly, 18–25% by the end of the acid hour, which the authors call "an expected outcome" for an enteric polymer with limited solubility below pH 6 R15.a–b·C23. After the shift to pH 6.5 they rose to 71–90% R15·C34. The HPMC E5 LV ASD released about 43% within 10 min of acid and peaked at about 81% within 50 min, then settled at about 75% after the shift; the authors find this "not surprising" because HPMC dissolves readily in acid R15.c·C36. The tablets stayed at about 15% in both stages R15.d–e.

Table C. Drug release (%) in the two-stage test, target 1.5 mg/mL, read from Figure 5 (approximate; mean of n = 3). Text values: HPMCAS-MF ASDs 18–25% in acid and 71–90% after the shift; HPMC-based ASD about 81% in acid and 75% after; tablets at most 15% in acid R15.

The authors describe the post-shift release of the HPMCAS ASDs, "up to 90%", as "a small decrease" from the complete or near-complete single-stage release C34. For the 20% DL ASD this holds (about 90% against about 100%), but the 30% DL ASD reached about 71% against 94% in the single-stage test, a gap of about 23 percentage points R15.a–b·R9.a–b·X7. The authors suggest the shortfall "might be attributable to a small extent of drug crystallization" of undissolved particles held in acid, a failure mode reported before for HPMCAS ASDs (Nguyen 2023, refs 74, 75), and add that "this speculation requires further investigation" C35. In Section 3.5.2 the authors also write that both tablets released faster than "the ASDs" in acid; in Figure 5 that holds only for the HPMCAS-MF ASDs during the first 30–40 min, and never for the HPMC E5 LV ASD C22·X6. Two smaller points: the methods give "an additional ∼30 min" after the shift, but Figure 5 shows 50 min; and the HPMC E5 LV loading appears only in the legend, with no single-stage test at that loading for comparison X8·X9.

Discussion

How the authors connect the results

The argument of the paper is a chain from solution behaviour to release. The polymer that keeps supersaturated PZQ from crystallizing longest is chosen as carrier; the ASD then dissolves to a supersaturated solution; whether that supersaturation survives decides whether release stays high C5·C7·C21. Where the polymer is a strong inhibitor for the form in the ASD, release is high; where it is weak, the drug "likely" crystallizes and release falls C20·C40.

The switch between the two outcomes is the inhibiting power of the polymer for the specific PZQ form, and here the paper offers three explanations, all hedged and none tested C26·C27·C28. First, the solid forms differ: RS-PZQ crystallizes as an anhydrate with all molecules in the syn conformation, R-PZQ only as hydrates with the anti conformation, and these conformational differences "could directly impact" how the polymers interact with the drug; the weaker inhibition "may be attributed" to them C26. Second, water bound to one carbonyl group of each R-PZQ molecule "may reduce the ability of the drug molecules to interact with the polymer" C27. Third, since chiral polymers such as HPMCAS and HPMC can discriminate between enantiomers (refs 65, 66), the authors find it "not unreasonable to speculate" that they interact more with S-PZQ and so protect the racemate better than R-PZQ C28. The paper did not identify which crystal form appears in the induction-time or release tests, nor measure drug–polymer interactions in solution X15.

The DL results are read through the same lens. With HPMCAS-MF, release falls gradually rather than collapsing, which the authors take as a sign of polymer-controlled release and as the absence of a "falling-of-the-cliff" effect for both drug forms C16·C17·C18. For the gastric stage the explanation is polymer solubility: HPMCAS-MF stays undissolved at pH 1.6 and holds the drug back, HPMC dissolves and releases it C23·C36. Because PZQ is taken with food, when gastric pH is higher, the authors judge that the slightly impaired post-shift release of HPMCAS ASDs "may not be of concern", while calling for studies of pH sensitivity C37.

What it suggests for formulation

The authors' practical conclusions are that HPMCAS-MF is the carrier of choice for R-PZQ ASDs, that HME is a suitable process for PZQ ASDs, and that ASDs of R-PZQ alone "may help overcome" the taste, size and compliance problems of the racemic tablet, with significant implications for commercial products C40·C32·C46. They "anticipate" improved pharmacokinetics and oral bioavailability, but no in vivo work was done C45.

Note by the AI writer (not discussed in the paper): the clearest lesson for formulation is that a polymer ranking made with the racemate did not transfer to the enantiomer. Both of the best inhibitors for RS-PZQ lost most of their effect with R-PZQ, and with HPMC E5 LV this was visible as lost release within an hour R4·R12.b. For an enantiopure product the inhibitor screen should be run with the enantiomer itself. In these data, HPMCAS-MF at 20–30% DL is the region where release is both high and reproducible; at 40% DL the RS-PZQ error bars widen R9·X13. HPMC E5 LV gives fast release at gastric pH, which may matter if release in the stomach is wanted, but it is the weaker partner for R-PZQ R15.c·R12.b.

Limitations

Limitations stated by the authors. The authors say that "the exact mechanism behind the improved release performance of enantiopure PZQ ASDs and their long-term physical stability was not determined", and that the mechanism of the improved R-PZQ release at 40% DL "requires further investigation" C41. They call their explanation of the two-stage shortfall a "speculation" C35, recommend further study of the pH sensitivity of release C37, note that HME of R-PZQ was not investigated C32, and state that pharmacokinetic studies are not reported C45.

The AI writer's own judgement (1): what the release comparisons can carry. The large difference between ASDs and tablets, roughly 83–117% against at most 16%, is direct and replicated, but it compares a whole tablet in 400 mL with sieved powder in 50 mL R8·R9·R14·X5. Several "complete" plateaus exceed 100% by up to about 17 points without comment, and the small differences built on top of them, such as HME against SE, are not resolved X3. At 1.5 mg/mL, RS-PZQ release went above its own amorphous solubility, so part of the measured drug may not be molecularly dissolved X4. The R-PZQ advantage at 40% DL lies within the racemate's error bar, and no claim of difference is backed by a statistical test X13·X12. The HPMC E5 LV 20% DL series at 1.5 mg/mL, low and highly variable, is not discussed X11.

The AI writer's own judgement (2): stability and mechanism. The statement that the ASDs of both polymers "remained physically stable for several months" conflicts with the reported crystalline peaks in the stored HPMCAS-MF ASD, and it rests on one 20% DL ASD per polymer X1·X2. The explanations for the weaker inhibition of R-PZQ (solid form, conformation, carbonyl–water binding, stereoselective binding to S-PZQ) and for the losses in release (crystallization from HPMC E5 LV, crystallization in acid) are untested; the crystalline solubility was measured on solids described as "in hemihydrate form" whose identity for RS-PZQ is not established, and no crystal formed during the tests was identified C26·C27·C28·C20·C35·X15. The induction-time comparison between R-PZQ and RS-PZQ was run at each drug's own amorphous solubility, so the absolute concentrations differed X14.

Conclusion

The paper shows, with direct in vitro measurements, that HPMCAS-MF and HPMC E5 LV are the strongest of seven polymers at delaying crystallization of supersaturated racemic praziquantel, that their ASDs release far more drug than generic or branded tablets in pH 6.5 buffer, and that with HPMCAS-MF an ASD of R-PZQ alone releases as well as the racemate at 30% DL and somewhat better at 40% DL R3·R8·R9·R14·R12·R13. It also shows that both polymers inhibit R-PZQ crystallization much less than RS-PZQ crystallization, and that with HPMC E5 LV this costs R-PZQ release R4·R12.b. The mechanisms behind these differences, the long-term stability of R-PZQ ASDs and their behaviour in vivo remain open, as the authors themselves state C41·C45. In the AI writer's view, if the R-PZQ result holds up in finished dosage forms and in vivo, an enantiopure ASD with HPMCAS-MF is a plausible route to a smaller, less bitter praziquantel product C46.