This study confirms previous data, that treatment of larger irradiation fields does not improve overall survival in stage I and II follicular lymphoma. Wilder et al. reported the M.D. Anderson experience of 80 patients treated with IFI (9%), regional RT (54%; treatment of 1–3 adjacent, grossly uninvolved nodal regions) or EFI (37%); TNI irradiation was not practiced . After a median follow-up of 19 years for living patients, no difference in OS was observed between EFI and IFI/regional RT: 49% versus 40% at 15 years. Recurrences were more frequently observed in patients treated with EFI; however, there were imbalances of stage distribution and total irradiation dose and no multivariate analysis was performed. Mac Manus et al. described the Stanford experience of 177 patients and differentiated between irradiation of the involved side or both sides of the diaphragm . After a median follow-up of 7.7 years, FFP at 10 year were 36% and 67% for treatment of the involved side only and both sides of the diaphragm, respectively, which was highly significant. However, this improvement in disease control did not transfer into increased OS.
In our own study, evaluating the impact of TNI and EFI on FFP and OS was difficult because of a correlation between the target volume concept and established prognostic parameters: age of the patients, performance status and stage of disease. No difference of FFP was observed between TNI and EFI in univariate and multivariate analysis. In agreement to Stuschke et al. , irradiation of uninvolved lymph node regions with a median dose of 30 Gy was effective in reducing out-of-field recurrences. However, an increased rate of in-field recurrences and extra-nodal out-of-field recurrence counterbalanced this beneficial effect of TNI. Regarding OS, patients treated with TNI were younger and had a better performance status and as a consequence, 15-years OS was 65% and 34% after TNI and EFI, respectively. However, only age and Karnofsky performance status and not the target volume concept remained statistically significant in multivariate analysis.
The 10-years OS of 64% for the patient population with stages I/II disease is considered as representative for early stage follicular lymphoma because our department is the only radiotherapy centre within a radius of minimum 50 – 100 km and rather strict referral patters in the geographical region; a potential patient selection bias is consequently limited. The observed OS is in good agreement with published data in the literature about primary radiotherapy for stages I/II follicular lymphoma: two recent literature reviews reported 10-years OS rates ranging between 46% and 79% (weighted average of 63% in Heinzelmann et al.) after primary radiotherapy using mostly IFI or EFI [11, 12]. No obvious correlation of OS and irradiated volume is seen in this literature review. This further supports the hypothesis of no OS benefit after TNI.
The larger treatment volumes of TNI came at the price of increased toxicity. While the first series of EFI irradiation was well tolerated with only 5% of the patients developing grade III toxicity, 16% of the patients suffered from acute toxicity grade ≥ III during the second series, which completed the TNI. Additionally, chronic toxicity grade II was observed more frequently after TNI (31%) compared to EFI (17%).
Randomized data about the question of the appropriate irradiation volume are missing. A German multicentre phase III trial, which randomized between TNI and EFI, closed in 2007 and final results are awaited. Until results of this trial are available, increased acute and chronic toxicity with no apparent benefit in terms of survival and disease control as observed in this study do not support the concept of TNI compared to EFI for stage I and II FL.
Twenty one patients with stage III FL were treated with radiotherapy alone and were included into this study. Unfortunately, prognostic factors were not fully balanced between stages I/II and stage III. Patients at stage III FL were younger indicating that age was a selection criterion for radiotherapy alone in stage III patients. As expected based on the increased lymphoma burden, LDH was more frequently elevated in stage III patients. TNI was the planned treatment in all stage III patients and was successfully completed in almost 90% of the patients.
Outcome was favourable with 10-years OS and FFP of 61% and 62%, respectively. OS and FFP were not significantly different to stages I/II. This favourable outcome is in good agreement with data in the literature [7, 8, 13, 14], where 10-years OS ranged between 47% and 65%. Only a small number of patients with stage III disease was at risk after 10 years in our study; however, the Stanford experience of 66 stage III patients treated with radiotherapy alone described no disease recurrence after 10 years and only 5/29 lymphoma deaths after 10 years . These data are promising especially in the light of the recently published British trial, which showed that radiotherapy doses for indolent lymphomas can be reduced without negatively affecting clinical outcome. Consequently, TNI with reduced irradiation may have the potential of long term disease control and survival with simultaneously decreased acute and chronic toxicity . Of course, such concepts need to be validated within clinical trials.
Some limitations of our study need to be discussed. The retrospective nature of this study is associated with well-known limitations. The decision making process between EFI and TNI considered well established prognostic factors like age, performance status and stage of FL, which made the statistical comparison of the two target volume concepts difficult. Additionally, the reconstruction of the decision making process in favour of radiotherapy and against chemotherapy in stage III FL could not be fully reconstructed: the ratio of stage I + II : stage III patients in the International Lymphoma classification project was 2:1  compared to 4:1 in our study, which suggests some selection process for radiotherapy.