The original 2013 molecular testing guideline regarding the selection of patients with lung cancer for treatment with targeted tyrosine kinase inhibitors was recently updated.1 The new International Association for the Study of Lung Cancer/College of American Pathologists/Association for Molecular Pathology guideline now amiably stipulates that any cytology sample may be tested if it has adequate cellularity and preservation.1 Although cell blocks were recommended over direct smears in the initial guideline, this updated version permits molecular testing, including next-generation sequencing, to be performed on any cytology preparation. This is not unexpected, because several studies have since provided ample evidence supporting the change; particularly that both archived and freshly prepared direct smears can serve as a reliable specimen source for molecular studies,2, 3 especially when cell blocks contain insufficient tumor cellularity. Moreover, nonformalin-fixed, paraffin-embedded cytology specimens may offer not only greater cellularity but also better quality material for molecular testing than cell blocks.4 Accordingly, in the near future, we are likely to witness more widespread use of cytology smears, and possibly also touch preparations, for molecular testing. If cytology smears are going to be used for molecular studies (eg, by scraping off valuable cellular material), then these slides will need to be sacrificed. Whole-slide imaging (the digitization of entire glass slides) offers a mechanism to immortalize and digitally archive selected slides that are destined to be destroyed for this purpose. To date, only limited pathology laboratories have embarked on scanning all of their cytology slides. If digitally archiving sacrificed cytology slides does become a mainstay in clinical practice, then laboratories will need to pay more attention to barcoding their smears to automate scanning as also might well consider z-stacking. Currently, there are no guidelines governing the retention of digital slides. Arguments against permanently storing digital slides are the increased cost of storage and potential future exposure to litigation. Fortunately, the cost of digital storage is likely to decrease over time because of cheaper and more cost-effective storage technologies. Some individuals advocate storing digital slides for only as long as they are required to retain glass slides. According to the College of American Pathologists checklist (anatomic pathology [ANP] 12500), digital images used for primary diagnosis must be retained for a period of 10 years if the original glass slides are not available. The cytology community is encouraged to advocate that digital cytology slides become a legally acceptable substitute for glass slides if they are appropriately scanned and stored. Unfortunately, many laboratories today rely on manual methods instead of digital tools to designate which regions on slides need to undergo molecular testing. For hematoxylin and eosin-stained cell blocks or similar tissue sections, this often entails designating areas of interest on the glass coverslip of a slide using a permanent pen marker. This is then used to guide subsequent manual macrodissection or scraping of tissue off the slide. However, when a single slide is available, manually dotting it may pose a problem, especially if the cytology and molecular laboratory are geographically separated. For this reason, molecular laboratories may have an extra hematoxylin and eosin recut made for themselves, which is wasteful of precious tissue. Scanning those glass slides upfront with digitally annotated areas of interest and subsequently sharing the digital slides with multiple users might offer a superior alternative. Now that cytology preparations are considered by the updated guideline to be acceptable for molecular testing in lung cancer, it is going to be increasingly important to evaluate the adequacy of this material. In addition to substantiating adequate cellularity, samples will need to contain enough tumor cells. An accurate, reproducible, and timely method is required to accomplish this. Visual estimation of tumor cellularity is subject to human perception bias and error. For this reason, image analysis should be considered to help better quantify and document sample cellularity in an objective way. Indeed, it has been demonstrated that using quantitative image analysis of cellularity before ordering molecular testing in digitized cytology material is feasible, more accurate, and less variable than manual counts.5 Automated image analysis also can help reduce the time pathologists spend on performing these tasks. The heterogeneity and scattered distribution of tumor and nontumor (eg, inflammatory and stromal) cells in cell blocks pose a challenge to manual selection of just the tumor cells. Laser capture microdissection (LCM) has been used to address this problem, allowing isolated tumor cells to be selected within sections. However, even with LCM, it can still be tedious and time consuming to manually select tumor cells. Fortunately, Jason Hipp and colleagues were able to integrate image-analysis software used to detect individual tumor cells with a LCM device.6 With this technology, after a pathologist identifies a tumor cell in a digital slide, the algorithm will automatically search the entire image for similar tumor cells so that they can be microdissected from the corresponding glass slide.7 This enables computer-guided LCM to be automated and provides a more rapid, hands-free, and scalable solution for molecular testing. Clearly, there are many ways in which digital pathology can be leveraged to improve the workflow in the cytology laboratory and the assessment of cytology material destined for molecular testing. These novel applications are likely to become progressively more important, especially if the new guideline truly does promote more widespread molecular testing of any cytology preparation with adequate cellularity. No specific funding was disclosed. Liron Pantanowitz serves as a consultant for Hamamatsu and serves on the Leica and Ibex medical advisory boards, outside the submitted work. Liron Pantanowitz, MD is a Professor of Pathology and of Biomedical Informatics at the University of Pittsburgh. He is the Vice Chair for Pathology Informatics and Director of the Pathology Informatics Fellowship at the University of Pittsburgh Medical Center (UPMC). He is also Director of the Cytopathology Division at UPMC Shadyside. Dr. Pantanowitz is an Editor-in-Chief of the Journal of Pathology Informatics. He is a member of the executive board of the American Society of Cytopathology, a member of the board of directors for the Digital Pathology Association, serves as a council member for the Association for Pathology Informatics, and is a member of the College of American Pathologists Digital Pathology Committee. He has widely published in the field of pathology informatics, including digital imaging and its application to pathology.