To the editor,

We thank Professors King and Ai for their insightful response to our systematic review and meta-analysis [1]. The research being performed by their team has been fundamental to much of our understanding of nasopharyngeal carcinoma (NPC) diagnosis, prognosis, and treatment.

The authors include an important and nuanced discussion of the grading system that they have developed for MRI diagnosis of NPC. This is of critical importance to radiologists’ interpretation and improvement of diagnostic accuracy. Unfortunately, many other papers that we included in our meta-analysis did not discuss their diagnostic criteria in depth. This precluded us from performing a feature-specific subgroup analysis. Use of reproducible imaging criteria will help radiologists to maintain consistent accuracy as pointed out in their letter.

We wanted to clarify that the paper by Liu and colleagues [2] was published during the peer-review process for our meta-analysis, and we adjusted our statistical analysis and manuscript to ensure its inclusion. Unfortunately, based on published information, concern about potential cohort overlap led us to exclude the earlier King and colleagues’ [3] study from our final analysis. Even allowing for this, MRI has been found to have a very high-test accuracy within the pooled cohort, and it is unlikely that inclusion of the further paper would alter the overall outcome.

We are very encouraged by the work being done by the team at CUHK and collaborators. In particular, they are developing a screening program which utilizes clinical data, serology, and imaging to facilitate early diagnosis of NPC. Multicenter, prospective validation of their MRI diagnostic criteria will be particularly useful given the role of imaging in this screening paradigm. It will also facilitate future comparative research and standardized reporting, both of which are essential to comprehensive cancer care, and for improving patient outcomes.