Two or more extranodal sites of disease identified by a PET/CT scan can predict a higher risk of secondary CNS involvement in patients with diffuse large B-cell lymphoma, Australian led research shows.
The single-item model of risk stratification provides an alternative to other validated tools such as the six-item CNS International Prognostic Index (CNS-IPI).
Dr Chan Cheah, a co-author of the multicentre, retrospective study of 1,532 patients, told the limbic all risk models suffer from limitations, particularly poor positive predictive values.
Yet it was important to find a balance between identifying those patients at high risk while minimising exposure for other patients to the potential side effects and toxicities of prophylaxis for CNS relapse.
“Data support the CNS-IPI for risk stratification and patients with a score of four or more should be considered for prophylaxis,” Dr Cheah said.
However, the current study also supported an alternate and simpler model to define patients at high risk of CNS recurrence.
“Among patients with >2 extranodal sites, 105/143 (73%) were also high risk according to the CNS-IPI,” the study said.
“The absolute number of extranodal sites is also a continuous rather than a dichotomous variable – as the number of sites increases, so does the risk,” Dr Cheah said.
Dr Cheah, a haematologist at the Sir Charles Gairdner Hospital in WA, said more work was required on biological markers of CNS involvement such as expression of MYC and BCL2 proteins and especially rearrangements of MYC.
More research was also required on the efficacy of various regimens for CNS prophylaxis.
He said while most of the prophylaxis is typically given after the sixth cycle of CHOP, up to 20% of patients develop secondary CNS involvement in the first six months after diagnosis.
“So not giving prophylaxis early risks missing some people,” he said.