We present in vivo experimental evidence that multispectral quantitative photoacoustic tomography (qPAT) has the potential to detect osteoarthritis (OA) in the finger joints. In this pilot study, two OA patients and three healthy volunteers were enrolled, and their distal interphalangeal (DIP) joints were examined photoacoustically by a multispectral PAT scanner. Images of tissue physiological/functional parameters including oxyhemoglobin, deoxyhemoglobin, oxygen saturation, and water content, along with the tissue acoustic velocity of all the examined joints, were simultaneously recovered using a finite element reconstruction algorithm for multispectral photoacoustic measurements. The recovered multispectral photoacoustic images show that the OA joints have significantly elevated water content, decreased oxygen saturation, and increased acoustic velocity compared to the normal joints.