I read with interest the paper entitled: Cell of origin is not associated with outcomes of relapsed or refractory diffuse large B cell lymphoma.
The paper is certainly well written and results are well reported. However, the therapeutic the paper refers to has largely changed. It may be that once Cart cells, Tafa, Lonca or Polatuzumab are implemented as second line (just to cite some) results may not be confirmed. In details:
1) I suggest that together with this historical cohort, authors add a series of patients similarly analyzed but with the novelty of newer drugs in first or second relapse. The impact of COO under new drugs is largely underinvestigated and would represent a major strengthness of the paper
2) In the abstract and Tables the % of DEL and DHL are wrong because they do not refer to analyzed patients. DEL should be 30 of 135 (and not 9%), and DHL 19 of 163 (and not 5%). Please verify and correct.
3) When addressing COO impact, I would comment on elderly patients that have not been treated with curative intent (i.e. >85 years?). A quick analysis relying on patients treated with curative intent (both upfront and later on) would be useful
4) I like on page 12 the detailed description of patterns of relapse and second line Tx
5) Although Hans algorithm represents the main standard, several other algorithms (ie Visco-Young) are quite used on the US and Europe, and recognized by WHO. Still if expression data are not available, I suggest to comment on them.