What Solid Form Solutions Does

Solid Form Solutions (SFS), now part of Cambrex, is a specialized pharmaceutical services company focused on a critical early-stage challenge: determining the optimal solid form of a new Active Pharmaceutical Ingredient (API). A drug's solid form—its crystal structure, polymorph, salt, or cocrystal—directly impacts its stability, bioavailability, manufacturability, and intellectual property. SFS provides expert analysis and experimental services to identify the most viable form, de-risking downstream development for its clients. With laboratories in East Rutherford, New Jersey, and a heritage dating to 1981, the company serves a global clientele of biopharma firms, leveraging techniques like X-ray diffraction, thermal analysis, and spectroscopy.

Why AI Matters at This Scale

As a mid-market player embedded in a larger corporation, SFS operates at a pivotal scale for AI adoption. It is large enough to generate significant, valuable datasets from thousands of client projects, yet agile enough to pilot and integrate new technologies without the inertia of a mega-corporation. In the highly competitive Contract Development and Manufacturing Organization (CDMO) and services sector, speed and predictive accuracy are key differentiators. AI offers a path to transform their core service from a labor-intensive, iterative experimental process into a more predictive, efficient, and high-value consultancy. For a company of 1,000-5,000 employees, strategic AI investment can create a defensible technology moat, allowing them to win more business and command premium pricing for accelerated insights.

Concrete AI Opportunities with ROI Framing

1. AI-Powered Polymorph Prediction: The manual screening for polymorphs is a major time and cost sink. An ML model trained on historical experimental data (e.g., solvent parameters, molecular descriptors, outcomes) can predict the most likely polymorphic outcomes for a new molecule. ROI: Reducing physical screening campaigns by 30-50% directly cuts labor and material costs, while accelerating project timelines, enabling the team to handle more client projects annually.

2. Computer Vision for Microscopy Analysis: Scientists spend hours analyzing crystal images for shape and size distribution. A convolutional neural network (CNN) can be trained to classify and quantify these features instantly. ROI: Automating this repetitive task frees senior scientists for higher-value analysis, improves consistency, and provides real-time feedback during crystallization process development, reducing batch failures.

3. Predictive Stability Modeling: Long-term stability studies are time-consuming and delay decisions. AI can correlate early-stage solid-state properties (e.g., hygroscopicity, surface area) with long-term stability outcomes from past projects. ROI: This allows for earlier, data-driven go/no-go decisions on formulation strategies, potentially shaving months off development schedules and preventing costly late-stage rework for clients.

Deployment Risks Specific to This Size Band

For a firm in the 1,001-5,000 employee band, key risks include resource allocation: competing capital and talent needs between core operations and speculative tech projects. Data silos may exist between legacy instrument databases and newer systems, requiring upfront investment in integration. There is also the specialist talent gap—finding individuals who understand both pharmaceutical solid-state science and data science is difficult and expensive. Finally, change management is critical; convincing seasoned scientists to trust and use AI-generated predictions requires careful change management and demonstrating unambiguous value on pilot projects to build internal credibility.