Low Dose Total Body Irradiation for Relapsed Low Grade Non-Hodgkin ’ s Lymphoma : Experience of National Cancer Institute , Cairo

Background and Purpose: The relapsed low grade non-Hodgkin’s lymphoma (LG-NHL) is currently incurable disease and the optimal treatment regimen has not determined yet. Low dose total body irradiation (LTBI) provides an alternative mechanism of action against cancer cells rather than direct cell kill. The mode of action of LTBI is immune-modulatory effect, induction of apoptosis and hypersensitivity to low radiation doses. The aim of our study is to evaluate the effect of LTBI on relapsed LG-NHL and reporting our experience at National Cancer Institute, Cairo (NCI, Cairo). Material and Methods: Fifty eight patients with relapsed LG-NHL and received LTBI studied retrospectively. LTBI dose was 1.6 Gy/8 fractions divided on 2 courses; each course 4 fractions treated over 4 days with 2 weeks rest between the 2 courses. Results: The median age is 54 years; 65% of the patients are men. Forty (69%) patients had performance status of 2 or more. Twenty seven patients were stage II/III and 31 patients (53%) had stage IV disease. Twenty six (45%) patients had bulky disease more than 10 cm and 22 (38%) patients had B symptoms at the time of relapse. The extranodal disease was present in 17 patients (29%) and 78% of the patients received >3 regimens of chemotherapy before referral to LTBI. Twenty three patients received IFRT (mean dose 32 ± 4 Gy) to initially bulky sites after LTBI. Fourteen patients (24%) achieved complete remission (CR) while 45%, 21% and 10% had partial remission (PR), stable disease (SD) and progressive disease (PD) respectively. The median PFS duration was 14 months and the median OS duration was 39 months. Stage VI, >3 regimen of chemotherapy and bad response to LTBI (SD) affected progression duration adversely (0.03, 0.05 and 0.01 respectively). The response to LTBI is the only factor afCorresponding author. Y. Bayoumi, A. Radwan 26 fected the OS duration significantly. The 3-year PFS was 19% ± 9%, and 3-year OS was 45% ± 8%. Stage IV was the only factor affected the 3-year PFS significantly with p value 0.03. The hematological toxicity was the main side effect of LTBI. Eleven patients developed G3/4 anemia while 8 patients only developed G3/4 thrombocytopenia and 13 patients developed G3/4 leucopenia. Conclusion: The use of LTBI in patients with relapsed low grade NHL is a feasible, effective and tolerable treatment that is worthy of testing in a future with chemotherapy and Rituximab maintenance.

The Low grade non Hodgkin Lymphoma (LG-NHL) represents about 35% of all malignant lymphomas [1].In general, these diseases are characterized by a slow clinical pace and by responsiveness to a variety of treatment approaches; relapse is essentially inevitable in advanced stage.In relapsed or refractory patients to initial therapy, the response to different chemotherapy combinations is transient and long-term survival is rarely seen [2].High-dose therapy with stem cell support may improve the response, but the impact on survival is not well established [3].
The use of Low-dose total body irradiation (LTBI) for treating LG-NHL started at the beginning of the twenties of previous century.The interest of using LTBI compared to total nodal and subtotal nodal irradiation is mainly attributed to the evidence of having different mode of action on the cancer cells rather than direct cell kill.Experimental data suggest that the mode of action could be explained by 3 mechanisms namely; induction of apoptosis [4], intrinsic hypersensitivity to low radiation doses [5] and immune-enhancement [6].
The evidence of clinical efficiency of LTBI in LG-NHL grows from the late sixties and early seventies with the use of very low individual TBI fraction sizes (0.1 -0.25 Gy) given several times a week until a cumulative dose of approximately 1.5 -2 Gy is reached [7]- [9].The objective of this study is to report the National Cancer Institute, Cairo (NCI, Cairo) experience in treating relapsed LG-NHL patients with LTBI.

Materials and Methods
The medical records of the patients with low grade NHL received LTBI at NCI, Cairo and Damieta cancer institute (DCI) were reviewed.The patients subjected to the staging work up as per the NCI, Cairo guidelines including CT neck, chest, abdomen and pelvis.Bone marrow and complete lab including differential blood count and chemistry were routinely done for all patients before LTBI protocol.
LTBI consisted of two cycles; each one of 0.8 Gy/4 fractions/4 days; 0.2 Gy per fraction.The 2 cycles separated by 2 weeks rest.The total dose is 1.6 Gy over 4 weeks.LTBI was given by AP/PA technique and the dose was calculated to the patient's midline at the level of the umbilicus using the patient translation and beam zone method [10].Dose homogeneity along the patient's midline was estimated to be ±7%.An alternative technique was used; the semi setting position with extended SSD technique of 4 meter distance at DCI with homogeneity ±15%.No lung shielding or compensators were implemented [11].
LTBI is given if WBCs ≥ 3.0 × 10 9 /L; platelets ≥ 100 × 10 9 /L and hemoglobin above 10 g/dl.The blood sample were taken before 1 st and 2 nd radiotherapy cycles of LTBI, and then after 2 weeks from the end of LTBI.Blood samples were taken in the patients who developed hematological toxicity on weekly basis till complete recovery.The best supportive care for the patients who developed haematological toxicity as per the NCI Cairo guidelines were offered.
Patients with bulky disease (10 cm) received additional involved-field radiotherapy (IFRT) to the initial bulky sites.The IFRT dose ranges from 25 to 35 Gy as per the field size and disease burden after LTBI and was given as daily 1.8 to 2 Gy per fraction, five times a week starting 4 -6 weeks after the end of LTBI.
Disease parameters and tumour responses were defined and evaluated according to the WHO criteria [12].Response evaluation was done after 2 month from the end of whole treatment.CR stated for complete remission, PR for partial remission, SD for stable disease and PD for progressive disease.The assessment of response was carried out by the same tools of staging work up.Patients were followed monthly for the first 2 years then every 3 months thereafter.

Statistical Analysis
The progression free survival (PFS) was calculated from the date of documented CR/PR to the date of first sign of progression.Overall survival (OS) calculated from 1 st LTBI fraction to the date of death or last follow-up.The following prognostic factors were studied; age, sex, B symptoms, Ann Arbor stage (II/III vs IV), involved field radiotherapy, performance status, response to LTBI (CR/PR vs SD/PD), extranodal presentation and the presence of bulky disease.
The estimate of survival was performed with the Kaplan-Meier method.The log-rank test was used to analyze survival differences among subgroups of patients.The paired T-test, Chi square and Fisher exact tests were used for testing proportion and to compare variables.Level of statistical significance is considered <0.05.Statistical software used was the SPSS for windows.

Results
Fifty eight patients proved to be LG-NHL out of 202 patients treated using LTBI during the period from 1997 till 2006 were retrospectively analyzed.The pathological subtypes include small cell lymphocytic lymphoma (13 patients), follicular lymphoma (33 patients), and marginal zone lymphomas and others (12 patients).
The patients' age ranged from 39 to 67 years with a median age of 54 years.There were 38 (65%) men and 20 (35%) women.Forty patients (69%) had a WHO performance status of 2. Eight patients (14%) were stage 2, 19 patients (33%) stage III and 31 patients (53%) had stage IV disease.Twenty six patients (45%) had bulky disease (nodal or extranodal) and B symptoms were present in 22 (38%) patients.The extranodal disease is present in 17 patients (29%).Twenty three patients received IFRT out of the 26 patients presented by bulky disease.The 3 patients skipped the IFRT due to development of PD during the LTBI course.The mean dose of IFRT was 32 ± 4 Gy.The patients received one to four different types of chemotherapy regimens and from 6 to 32 courses before referral to radiation oncology department.Patients' characteristics are summarized in Table 1.
The follow up duration ranges from 10 to 73 months (mean 42 ± 17 month).The median PFS duration was 14 months and the median OS duration was 39 months.The effect of different prognostic factors on survival is illustrated in Table 2.
Stage, number of chemotherapy regimens and response to LTBI affected progression free duration significantly (0.03, 0.05 and 0.01 respectively).The response to LTBI is the only factor affected the OS duration significantly; the CR/PR patients showed 49 month median survival compared to 28 month for SD patients (p value 0.05).However the data should be taken cautiously due to the inhomogeneous distribution between the groups of patients.
The 3-year PFS was 19% ± 9% as shown in Figure 1 and 3-year OS was 45% ± 8% as shown in Figure 2. The effect of disease bulk, IFRT and stage were studied.The other factors excluded due to inhomogeneous distribution and small number of patients (Table 3).The stage before LTBI is the only factor affected the 3-year PFS significantly with p value 0.03.

The Toxicity of LTBI
The hematological toxicity was the main side effect for LTBI.Toxicity grading was done according to the Common Toxicity Criteria as shown in Table 4.The LTBI treatment is tolerable with no acute side effects that affect completion of the planned treatment schedule.Eleven patients developed G3/4 anemia while 8 patients only developed G3/4 thrombocytopenia and 13 patients developed G3/4 leucopenia.The median duration for leucopenia, anemia and thrombocytopenia was 5, 9 and 12 weeks respectively.No patients received growth factors and only 5 patients received blood elements transfusion.

Discussion
It is interesting that many new chemotherapeutic agents and monoclonal antibodies had failed to eradicate an apparently sensitive disease; low grade lymphoma.Even the median survival did not improve in the last 3 dec-      ades without any superiority of bone marrow transplantation regimens at least for refractory cases.Most of the studies using new chemotherapeutic drugs for relapsed indolent NHL showed progression free survival at 5 years ranges from 20% to 40% [13]- [15].The LTBI is a well known historical method of treating the relapsed LG-NHL.Yonkosky et al., treated 9 relapsed NHL with fractionated LTBI (1.2 -1.8 Gy) and reported CR rate more than 50% which is higher than our data that showed CR rate 24% only [16].Also, Jacobs and King randomized 108 patients with indolent NHL and CLL between CP and LTBI.The CR rate for LTBI group (n = 54), 52% and median survival was 57 months [17].
Apart from the proved equal effect of LTBI and chemotherapy in Jacobs and King study, the better CR rate in LTBI arm in their study compared to our cases is mainly attributed to the difference between the 2 the 2 studies in patient characteristics, as we targeted the relapsed and refractory cases while Jacobs and King targeted the new and patients in 1 st relapse.Also in our study; 45 out of 58 patients received 3 lines or more of chemotherapy and 40 patients out of 58 patients had bad performance status (2 or more).
There are some similarities between the results of our study and the one by Carabell et al. who treated 58 patients with stage III or IV by a total dose of 1.5 Gy/10 fractions; 2 fractions per week.Survival at 8 years was 52% with 14% relapse free survival compared to 45% and 19% respectively in our group of patients.The median PFS duration was 14 month in our data compared to 24 month for Carabell data.[18].
Meerwaldet treated 44 low grade lymphoma patients by LTBI (1.5 Gy) and 40 patients by CHVmP.The 5year PFS was 15% for LTBI compared to 19% in our study and the 5-year OS was 45% for (LTBI) which is similar to our study.Surprisingly enough, the chemotherapy (CHVmP) arm had PFS 25% which is not far beyond the results of LTBI in the same study [19].
Studying the prognostic factors in our patients received LTBI proved that stage is the only factor affected the 3-year PFS significantly with p value 0.03.
The median duration for progression affected significantly by stage also in addition to number of chemotherapy regimen given before referral to LTBI and response to LTBI.
The response to LTBI is the only factor affected the OS duration significantly; the CR/PR patients showed 49 month median survival compared to 28 month for SD patients (p value 0.05).These results are comparable to previously published studies addressing the same issue.Lybeert et al. showed that prior treatment, stage and sex had no influence on survival.Age was reversibly correlated with survival but not with PFS.On the other hand, low grade lymphoma patients had a significantly better PFS if they received LTBI as a primary therapy but survival was not significantly influenced [20].
In another study, previous treatment decreased PFS by univariate analysis (p = 0.066).Multivariate analysis identified the best indicators of response rate to be histology, and marrow involvement.The best indicators of PFS were histology, and TBI dose.The best indicators for survival were age histology and TBI dose [21].
Even one of the widely used therapeutic agent in relapsed indolent lymphoma-Rituximab-having 40% to 50% response rate [23] [24] which is comparable to the LTBI results.The longer response duration of Rituximab could be justified by the idea of maintenance therapy in most of trials of Rituximab.
Three randomized, prospective studies randomized relapsed LG-NHL to rituximab maintenance vs none after retreatment with combination chemotherapy including Rituximab in the initial treatment; all trials showed prolongation of response duration and median duration of PFS (43 months vs. 15 months) in favor of Rituximab maintenance [25]- [27].These data are still comparable to our data, as the PFS duration was 15 month in the arm without Rituximab maintenance compared to 14-month in our study.
The overall response rate in our study is around 69% however; there is clear evidence in the literature that combining LTBI with chemotherapy gave better results.The results of prospective pilot trial combining LTBI (100 -150 rads) with chemotherapy (CVP or C-MOPP) for previously untreated, stage II-IV, low-grade NHL patients [28] showed good results in terms of overall response (86% -90%) and complete remission rates (60% -80%).Also combination of LTBI and Prednimustine were evaluated in relapsed low-grade NHL patients.Six to nine courses of Prednimustine were given 2 months after a 1.5 Gy LTBI as consolidation therapy.The overall response rate was around 85%.The complete remission rate occurred in 12.5% and 24.30% in the CLL and LG-NHL patients respectively [29].
The philosophy of using immune-modulators and maintenance therapy in relapsed LG-NHL is gaining popularity in the last decades.Currently, the immune-modulatory effect of LTBI is well known as mentioned previously [6] and proved by NCI, Cairo publications in NHL and metastatic renal cell carcinoma cases [30]- [32].

Conclusions and Recommendations
LTBI, however it is underutilized, it is tolerable and effective palliative treatment for relapsed LG-NHL even for refractory and resistant cases with bad prognostic factors.The effect is comparable to chemotherapy regarding the response rate and progression free survival.
Based on its efficacy and different mode of action against cancer cells, LTBI can be combined to chemotherapy and Rituximab maintenance and tested in multi institutional prospective study.

Figure 1 .
Figure 1.Progression free survival for 58 patients with relapsed low grade non-Hodgkin's lymphoma received low dose total body irradiation.

Figure 2 .
Figure 2. Overall survival for 58 patients with relapsed low grade non-Hodgkin's lymphoma received low dose total body irradiation.

Table 1 .
Patient characteristics and prognostic factors.

Table 2 .
Median survival duration and the effect of different prognostic factors.

Table 3 .
3-year progression free survival and overall survival for the patients with the effect of homogenously distributed prognostic factors.