Surgery for rectal cancer is associated with significant side effects including wound infections, incontinence, sexual and bladder dysfunction, and long-term ostomies. Though studies have shown that patients who completely respond to preoperative treatment can safely avoid surgery, nonoperative options remain limited by the poor performance of MRI and endorectal ultrasound after initial therapy. Therefore, new imaging modalities are needed to improve posttreatment tumor assessment and enable the widespread adoption of nonoperative management in rectal cancer. An acoustic resolution photoacoustic microscope (AR-PAM) was constructed with high frequency ultrasonic transducer and near infrared laser. We performed initial phantom, and then imaged ex vivo human colorectal specimens to evaluate different AR-PAM characteristics in each tissue type (normal, untreated tumor, and treated tumor). Our data suggest that photoacoustic imaging can differentiate the distorted vasculature of rectal tumors from normal vascular patterns. However, the vascular distribution of rectal tissue in pathological complete responders showed similar distribution as the normal colorectal tissue; mucosa, submucosa and muscle layer are clearly presented in ultrasound images, while photoacoustic images have revealed that most vasculatures distribute in submucosa. Encouraged by these initial results, we are developed a high-speed scanning (1 second for 20mm B-scan) AR-PAM with laser pulse repetition rate of 1kHz for large field 3D imaging. Lateral resolution of 65μm, axial resolution of 45μm, and 8mm tissue imaging depth can be achieved.
Colorectal cancer is the second leading cause of cancer death in the United States. According to American Cancer society, the overall lifetime risk of developing colorectal cancer is about 4.7% for men and 4.4% for women. We have developed a rigid, endoscopic photoacoustic microscopy (PAM) probe for imaging of in vivo human colorectal cancers. In order to accommodate colon sections with different size (typically from 50 to 70mm), our 10mm diameter rigid probe uses an off-optical-axis, external mechanical scanning mechanism with a speed of 35deg/s instead of an on-optical-axis, internal mechanical scan mechanism. 532-nm pulsed laser light enters the ridged probe through a photonic crystal single mode fiber before it is collimated and refocused by a water-immersed objective lens onto the colon surface. A focused ultrasound ring transducer (40.5 MHz, 6.5mm focal length) receives photoacoustic signal from chromophores excited by laser beam. Imaging system performance specifications including resolution (6μm) and signal-to-noise ratio are quantified and verified from phantom imaging tests. Ex vivo human colon samples are studied to reveal microscopic features of normal colon, benign polyps, adenocarcinoma and cancer.