SAN DIEGO-“Currently, there are no standard response criteria for non-Hodgkin’s lymphoma, as there are for solid tumors, chronic lymphocytic leukemia, and Hodgkin’s disease,” Anonio J. Grillo-López, MD, said at the annual meeting of the American Society of Hematology.
SAN DIEGOCurrently, there are no standard response criteria for non-Hodgkins lymphoma, as there are for solid tumors, chronic lymphocytic leukemia, and Hodgkins disease, Anonio J. Grillo-López, MD, said at the annual meeting of the American Society of Hematology.
In non-Hodgkins lymphoma, in effect, a complete response (CR) is whatever the investigator decides it is, and this may vary from study to study, he said. The result is that a drugs purported ability to induce complete remission may double or even triple, depending on how CR is defined.
Dr. Grillo-López, of IDEC Pharmaceuticals, has long argued that standard response criteria are needed, and in an ASH poster presentation, he showed why. Analyzing data from IDECs own clinical trials of rituximab (Rituxan), he demonstrated that it would bepossible to report CR rates of 6% to 28%, using the same patients and with the same overall response rate of 48% (see Table 1). The difference depended solely on how normal lymph node size was defined.
In consultation with the FDA, we assembled a group of lymphoma experts who drew up criteria for us to use in our trials of rituximab, Dr. Grillo-López said in an interview. For CR, these include no evidence of disease, no palpable nodes, no nodes larger than 1 cm × 1 cm on CT scan, no new lesions, reconfirmation of the response at 28 days, no decrease in performance status, negative bone marrow (if bone marrow was initially positive), and normal liver and spleen.
The IDEC criteria also define duration of response as the interval between the date of the first observation of 50% or greater tumor shrinkage and the date of disease progression.
The response criteria were applied by a third-party, blinded, panel of NHL experts to the 166 patients in the rituximab pivotal study. The database includes the size of all measurable lesions, allowing a computerized evaluation of different sets of criteria.
We performed an analysis applying our rigorous criteria for normal lymph node size (1 cm × 1 cm) as well as two alternative criteria. As anticipated, our rigorous criteria resulted in a lower CR rate than less stringent criteria. The overall response rate was not affected, Dr. Grillo-López said.
Confirmation at 28 Days
The stringent criteria required a confirmatory evaluation at 28 days or later after the first evidence of a response. When this criterion was not applied, a progressive increase in CR rates was seen. Without confirmation, CR doubled, he said (Table 2 ).
The data review also found that the CR rates and response durations were not affected when only the six largest lesions (rather than all measurable lesions) were measured to determine CR. There is no reason to restrict study entry according to lymph node size, as this did not affect the response rate when our rigorous criteria were applied, he added.
The results suggest that literature reports of the efficacy of NHL anticancer agents must be viewed with caution in studies without clearly defined response criteria, he said. The NCI is currently developing standard criteria for measuring tumor size and classifying responses in NHL.