While the shift away from routine CLND to nodal surveillance for SLN+ melanoma has been rapid compared with the often tepid pace of evidence-based practice implementation, development of institutional processes to support and optimize nodal surveillance in real-world settings is ongoing. Our prior work demonstrated that nodal basin U/S, which are required at frequent intervals for 5 years after surgery, can vary widely in their quality and specificity for evaluating MSLT-II criteria for nodal recurrence.6,7 Following implementation of a collaboratively-designed synoptic template for melanoma nodal U/S at our large tertiary cancer center, significant improvements were seen in both the number of MSLT-II criteria reported and surgical provider and radiologist satisfaction with the clinical utility of nodal U/S reports.
While equivalent melanoma-specific survival between nodal surveillance and CLND was achieved in the closely monitored trial setting of MSLT-II, barriers to the delivery of consistent, high-quality nodal surveillance present challenges to achieving similar results in real-world settings. One such barrier includes loss to follow-up or difficulty achieving consistent adherence to ultrasound surveillance, which has been previously reported as ranging from 34% in a single-center study up to 83% for the highest-adherence center in a multi-institutional study (study range 35–83%, overall 58% U/S adherence).3,11 These rates are likely reflective of an early implementation period for these new surveillance practices, as evidenced by the improved adherence rates seen in the cohort presented in this study. Specific to nodal surveillance U/S, concerns over the quality and interpretability of nodal U/S reports have been previously identified. In a qualitative study from Mott et al. in which surgical and medical oncologists were interviewed, multiple participants reported the perception that radiologists may not be aware of MSLT-II criteria or the reasoning behind melanoma nodal surveillance and thus may (unknowingly) omit the melanoma-specific clinically relevant findings.7 Given the frequency with which SLN+ patients are undergoing nodal surveillance in lieu of CLND, as high as 80–96% of SLN+ patients in multiple post-MSLT-II multicenter studies,3,4,5 interventions to address this issue could benefit a significant proportion of patients with SLN+ melanoma.
After considering these concerns about reliability and consistency with the reporting of MSLT-II criteria, a synoptic reporting template seemed an ideal choice for improving melanoma nodal U/S reports when this need was identified at our institution.6 Synoptic reporting tools have become increasingly common across the spectrum of cancer care, including radiology synoptic reporting tools, such as the Breast Imaging Reporting and Data System (BI-RADS) and Thyroid Imaging Reporting and Data System (TI-RADS), as well as synoptic operative reports and pathology reports.12,13,14,15 Across these different disciplines, the primary benefits of the synoptic format have included improved communication between users (e.g., pathologist or radiologist and surgeon or medical oncologist), clearer recommendations for next steps, and improved documentation of relevant information for future clinical or research use.16,17,18,19 Results from this study demonstrate similar outcomes with a nodal U/S synoptic reporting template, which enabled surgeon/surgical APP survey respondents to identify relevant findings and understand appropriate next steps in management. Furthermore, radiologist respondents expressed greater confidence in making clinical recommendations using the clearly delineated MSLT-2 criteria without imposing additional burdens on their time. As the population of SLN+ melanoma patients undergoing nodal surveillance continues to expand post-MSLT-II, having succinctly communicated findings from the radiology to surgical teams should enable the consistent delivery of high-quality surveillance care and represents a primary benefit of this synoptic template.
Facilitating factors to the implementation of this synoptic reporting template included having surgeon and radiologist “champions” who were recognized leaders in their division/sections and who participated in the design of the synoptic template, solicited feedback, and encouraged use of the synoptic template. This collaborative team also created a consistent line of communication between departments that aided in identifying issues and developing solutions in real time. In terms of developing the synoptic template, aligning the clinical recommendations to a point-based system with the same criteria used in the MSLT-II trial (e.g., if 2 or more U/S criteria were seen, biopsy is recommended) helped ground the radiologists’ recommendations in the evidence source that the surgeons were most familiar with and already using to guide their nodal surveillance practices.
Several logistical barriers were identified during the initial implementation, some of which have already been addressed in subsequent PDSA cycles. As many of the patients receiving nodal U/S in the post-implementation period had undergone previous nodal U/S as part of their nodal surveillance, occasionally the prior non-synoptic reports would be “copy-forwarded” and updated with current findings but without the synoptic reporting language. This was addressed through education with the radiology sonographers and a custom addition of the synoptic template to the “menu” of interpretation templates available for soft tissue U/S studies, so that the synoptic template could be auto-imported into the text of the report draft. This mechanism also helped to address issues with use of the synoptic report across different anatomical nodal basins; sometimes a neck U/S performed for melanoma nodal surveillance would focus on evaluation of the thyroid, for example, and less on the cervical nodal basin of interest. During the implementation period, stakeholders also identified a need to locate and scan the sentinel lymph node biopsy scar. This was especially pertinent for head and neck melanomas because of their variable drainage patterns to areas, such as occipital and post-auricular basins, which may not routinely be included in a cervical nodal U/S that focuses on anterior (level I-V) nodes. Further iterative improvements to the use of the synoptic report are ongoing, as well as efforts to collaborate with other institutions to share the synoptic reporting template.
Limitations of this study that may affect its interpretation and generalizability include its single-institution retrospective design, as well as barriers and facilitators to implementation of the synoptic reporting template that may be institution-specific. Of note, while U/S criteria for nodal recurrence were specified by the MSLT-II study protocol for participating sites, there was no requirement for synoptic or other external reporting of U/S criteria during MSLT-II; therefore, comparison of findings from this U/S protocol with patterns of nodal recurrence has not been specifically studied in a randomized, controlled fashion. Additionally, while findings represent the early post-implementation period, this intervention will need to be evaluated over a longer period of time in future work to ensure its continued success and identify ongoing opportunities for improvement. While this strategy was feasible at our institution based on the current radiology workflow and the support of departmental project champions, we recognize that adaptations may be needed for other institutions to implement a similar intervention.
Comments (0)