Background: Early adjustment of treatment may benefit the patient. In order to guide treatment adjustment, use of early response (ER) or early complete response (ECR), judged after the few initial cycles of chemotherapy, is common in pediatric and also adult Hodgkin's and non-Hodgkin's studies. Paradoxically, almost no data support this strategy.
Patients and methods: The influence of ECR on outcome was evaluated in three series of advanced Hodgkin's disease (HD), leading to a series of questions.
Results: The 1982 EORTC study assessed prospectively the time frame needed to reach an apparent complete response (CR) through repeated tumor measurements. In patients assessed at mid-treatment before the fifth cycle, both 15 year freedom from progression (FFP) and overall survival (OS) were superior in ECR patients compared with other patients continued on the same treatment (61% versus 37%; P < 0.001). A series of questions arise from these observations. Question 1: is the shortening of treatment detrimental? In a randomized Swedish trial, in one arm treatment was shortened in patients evaluated from the fifth cycle as ECR as compared with the standard eight cycles arm, 10 year cause-specific-survival (CSS) was 53 versus 69% [not significant (ns)]; 10 year OS 49% versus 58% (ns). Conversely, in the EORTC 20884 study, ECR patients given only six cycles did as well as patients entering CR later and, for this reason, given eight cycles (identical 6 year event-free survival 75%). Question 2: is early treatment adaptation in patients who failed to reach ER beneficial? In the French MDH 90 trial, 15% of children failed to reach ECR after four cycles; in these children only, anthracyclines plus alkylating agents were given and the dose of radiotherapy increased, improving the results observed in the previous trial. In the EORTC 20884 study, patients who failed to reach an ECR were switched earlier to involved field RT: their results matched those of ECR patients, at the difference of the previous trial. Question 3: is ER a predicting factor that can be used with any type of treatment? Probably not, based on the German Hodgkin's Lymphoma Study Group trial HD 9: ECR is highly dependent on specific interval from treatment start and on treatment intensity.
Discussion: More general questions stem from these results. Question 4: is the definition of ER secured? With conventional imaging, the different methods for response assessment at end treatment also lead to different response rates; the assessment in the middle of treatment itself and the use of newer imaging techniques may further increase the variation. Indeed, question 5 is: is ER a concept based on any biology? Correlation to markers, 99mTc uptake, PET and hematological tolerance might help to pinpoint how and why ER represents a surrogate for final outcome.
Conclusion: ER is a surrogate for final outcome, reflecting both tumor burden and activity. This predictability may, and possibly should, impact on treatment.