Dual Saturation Index Competition Predicts LLPS vs. Crystallization Pathway Switching in Ionizable Drug ASD Dissolution

When a drug company makes a pill, they often use a trick called an amorphous solid dispersion — essentially freezing a drug in a glassy, disordered state so it dissolves faster than it would in its natural crystalline form. The catch is that once this glassy drug hits the fluids of your digestive tract, it can do one of two tricky things: it can either form tiny liquid droplets suspended in the fluid (a process called liquid-liquid phase separation, or LLPS), or it can snap back into crystals — and crystals dissolve much more slowly, potentially ruining the whole point of the pill. Predicting which path the drug will take, and when, is a major headache for pharmaceutical scientists. This hypothesis borrows a computational tool originally designed to track whether volcanic minerals will dissolve or precipitate in groundwater. The idea is to simultaneously calculate two 'saturation indices' — essentially two separate scorecards measuring how badly the drug 'wants' to become droplets versus how badly it wants to become crystals. Whichever score is higher wins, and that phase forms first. The twist for drugs is that many pharmaceutical compounds are 'ionizable,' meaning they gain or lose electrical charge depending on the pH of the surrounding fluid — and stomach acid (pH ~1.2) is very different from intestinal fluid (pH ~6.8). This pH sensitivity shifts the two scorecards relative to each other, explaining why the same drug might form harmless droplets in your intestine but do something completely different in your stomach. The hypothesis was tested against real experimental data for three drugs across four pH conditions, and it correctly predicted the observed behavior in 9 out of 12 cases — including correctly calling that posaconazole, an antifungal drug, forms droplets first in intestinal conditions but skips both behaviors in stomach acid. That's not perfect, but it's good enough to suggest the underlying logic is sound.

This is an AI-generated summary. Read the full mechanism below for technical detail.