Multiphoton microscopy: a potential "optical biopsy" tool for real-time evaluation of lung tumors without the need
Manu Jain1, Navneet Narula, Amit Aggarwal
1From the Departments of Urology (Dr Jain), Pathology and Laboratory Medicine (Drs Narula and Shevchuk), Biochemistry (Drs Aggarwal and Mukherjee, Mr Sterling, and Mr Salamoon), Thoracic Surgery (Drs Stiles and Altorki), and Surgery (Mr Chandel), Weill Cornell Medical College, New York, New York; and the School of Applied and Engineering Physics, Cornell University, Ithaca, New York (Dr Webb). Dr Aggarwal is now with the Department of Science, Borough of Manhattan Community College, New York.
Context:
Multiphoton microscopy (MPM) is an emerging, nonlinear, optical-biopsy technique, which can generate subcellular-resolution images from unprocessed and unstained tissue in real time.
Objective:
To assess the potential of MPM for lung tumor diagnosis.
Design:
Fresh sections from tumor and adjacent nonneoplastic lung were imaged with MPM and then compared with corresponding hematoxylin-eosin slides.
Results:
Alveoli, bronchi, blood vessels, pleura, smokers' macrophages, and lymphocytes were readily identified with MPM in nonneoplastic tissue. Atypical adenomatous hyperplasia (a preinvasive lesion) was identified in tissue adjacent to the tumor in one case. Of the 25 tumor specimens used for blinded pathologic diagnosis, 23 were diagnosable with MPM. Of these 23 cases, all but one adenocarcinoma (15 of 16; 94%) was correctly diagnosed on MPM, along with their histologic patterns. For squamous cell carcinoma, 4 of 7 specimens (57%) were correctly diagnosed. For the remaining 3 squamous cell carcinoma specimens, the solid pattern was correctly diagnosed in 2 additional cases (29%), but it was not possible to distinguish the squamous cell carcinoma from adenocarcinoma. The other squamous cell carcinoma specimen (1 of 7; 14%) was misdiagnosed as adenocarcinoma because of pseudogland formation. Invasive adenocarcinomas with acinar and solid pattern showed statistically significant increases in collagen. Interobserver agreement for collagen quantification (among 3 observers) was 80%.
Conclusions:
Our pilot study provides a proof of principle that MPM can differentiate neoplastic from nonneoplastic lung tissue and identify tumor subtypes. If confirmed in a future, larger study, we foresee real-time intraoperative applications of MPM, using miniaturized instruments for directing lung biopsies, assessing their adequacy for subsequent histopathologic analysis or banking, and evaluating surgical margins in limited lung resections.
