Introduction
Disease relapse is still an important outcome for patients treated for diffuse large B-cell lymphoma (DLBCL). Almost one quarter of patients will have disease progression after the completion of the first treatment and will be treated for relapsed disease [1], and treatment is not always curative [2]. The standard of care for these patients, which is high-dose chemotherapy followed by autologous stem-cell transplantation (ASCT), with or without irradiation, is not feasible for a large portion of patients [3], even in controlled trials.
In this setting, radiotherapy can be used either for local control or as a palliative treatment. The role of radiation in the standard treatment for these patients was established almost thirty years ago [3], in the first PARMA trial. In addition, radiation has been investigated in the setting before [4] and after [5, 6] ASCT, with good results. Even after other important randomized trials that omitted radiotherapy, such as the CORAL trial [7], the use of radiation has been recurrently investigated, since that trial presented high rates of relapse in sites that would have been treated with radiotherapy.
This study was designed to assess the use of radiotherapy for relapsed DLBCL in a single university hospital and to describe the population for whom this treatment has been favored.
Materials and methods
This was a retrospective study of patients diagnosed with DLBCL and treated with radiotherapy between July 2010 and July 2017 who presented recurrent disease. Patients that did not receive second line/salvage treatment were excluded.
Demographic and treatment characteristics were assessed. Demographic variables included age, performance status (Eastern Cooperative Oncology Group — ECOG scale), stage, the presence of bulky disease, B symptoms, extranodal disease and human immuno-deficiency virus (HIV) status. The revised International Prognostic Index (R-IPI) was applied and updated in all patients. Treatment variables included the use of radiotherapy in second-line treatment, doses and fields, second-line chemotherapy, autologous stem-cell transplantation (ASCT) as consolidative treatment and the use of total body irradiation (TBI) as a conditioning agent. Afterwards, patients were divided in two groups for subsequent analyses: one including those patients that received radiotherapy for the recurrence, and the other including those who did not.
Overall survival (OS) was assessed from the date of diagnosis. Second progression-free survival (PFS2), considered as any recurrence after salvage treatment, was assessed from the date of first progression to the date of second progression.
Statistical analysis consisted of descriptive, and frequencies analyses, with comparisons between groups by the Fisher’s Exact test. For the oncological outcomes and survivals, the Kaplan-Meier method was used with the Log-rank test for univariate analysis. Significance was set at 5% (p ≤ 0.05).
Results
In the studied period, 359 patients were retrieved and, after charts evaluation, 65 (18.1%) presented disease progression, but only 62 (17.3%) that received further treatment were included in the study. Most patients were male (61.3%) and mean age at first diagnosis was 56.7 years. The most common chemotherapy regimens used as second line were ICE (ifosfamide, carboplatin, and etoposide) and IVAC (etoposide, cytarabine, ifosfamide, mesna and methotrexate). Table 1 and Table 2 describe, respectively, demographic and treatment characteristics. The respective characteristics and comparisons of the two groups: one that received radiotherapy as part of second line treatment and the other that did not are presented in Table 3. The use of radiation in the second-line setting was correlated only to the use of chemotherapy regimens other than R-CHOP (rituximab plus cyclophosphamide, doxorubicin hydrochloride, vincristine, and prednisone) in the first-line treatment (p = 0.020).
|
Variable |
Number (n) |
(%) |
|
Age at diagnosis (years), mean (range) |
56.7 (24–72) |
|
|
Age (at diagnosis) |
||
|
< 60 |
36 |
58.1 |
|
≥ 60 |
26 |
41.9 |
|
Sex |
||
|
Male |
38 |
61.3 |
|
Female |
24 |
38.7 |
|
ECOG performance status at diagnosis |
||
|
0–1 |
51 |
82.3 |
|
2–4 |
11 |
17.7 |
|
Stage (Lugano system at first presentation) |
||
|
1 |
4 |
6.5 |
|
2 |
7 |
11.3 |
|
3 |
10 |
16.1 |
|
4 |
41 |
66.1 |
|
R-IPI (at diagnosis) |
||
|
Very good |
2 |
3.2 |
|
Good |
22 |
35.5 |
|
Poor |
38 |
61.3 |
|
Bulky disease |
||
|
Present |
48 |
77.4 |
|
Absent |
14 |
22.6 |
|
Extranodal disease |
||
|
Present |
52 |
83.9 |
|
Absent |
10 |
16.1 |
|
B symptoms |
||
|
Present |
49 |
79.0 |
|
Absent |
13 |
21.0 |
|
HIV |
||
|
Negative |
56 |
90.3 |
|
Positive |
6 |
9.7 |
|
Variable |
Number (n) |
(%) |
|
First-line |
||
|
First-line chemotherapy regimen |
||
|
R-CHOP |
50 |
80.6 |
|
Others |
12 |
19.4 |
|
Number of chemotherapy cycles in the first line |
||
|
≤ 6 |
24 |
38.7 |
|
> 6 |
38 |
61.3 |
|
Toxicity to first-line chemotherapy |
||
|
0–3 |
41 |
66.1 |
|
4 |
21 |
33.9 |
|
Response to first-line chemotherapy |
||
|
Complete response |
15 |
24.2 |
|
Partial response |
42 |
67.7 |
|
Unidentified |
5 |
8.1 |
|
First-line consolidative RT technique |
||
|
Mantle and/or inverted-Y |
2 |
3.2 |
|
Involved-field RT |
10 |
16.1 |
|
Involved-site RT |
14 |
22.6 |
|
Bulky disease or PR site only |
36 |
58.1 |
|
Second-line |
||
|
Second-line chemotherapy |
||
|
Cytarabine-based |
48 |
77.4 |
|
Others |
14 |
22.6 |
|
Second-line RT |
||
|
Yes |
15 |
24.2 |
|
Yes, as consolidative treatment |
5 |
- |
|
Yes, and as re-irradiation to primary site |
6 |
- |
|
No |
47 |
75.8 |
|
Second-line RT dose [Gy] |
||
|
30 |
8 |
53.3 |
|
> 30 |
7 |
46.7 |
|
ASCT as consolidation for second-line chemotherapy |
||
|
Yes |
4 |
6.5 |
|
No |
58 |
93.5 |
|
Patients characteristics |
Second-line RT |
p |
|
|
No N = 47 (75.8%) |
Yes N = 15 (24.2%) |
||
|
Age < 60 years > 60 years |
25 (53.2%) 22 (46.8%) |
11 (73.3%) 4 (26.7%) |
0.169 |
|
ECOG 0–1 2 or more |
38 (80.9%) 9 (19.1%) |
13 (86.7%) 2 (13.3%) |
0.608 |
|
R-IPI Very good Good Poor |
2 (4.3%) 15 (31.9%) 30 (63.8%) |
0 7 (46.7%) 8 (53.3%) |
0.080 |
|
Bulky disease Absent Present |
12 (25.5%) 35 (74.5%) |
2 (13.3%) 13 (86.7%) |
0.325 |
|
Extranodal disease Absent Present |
8 (17.0%) 39 (83.0%) |
2 (13.3%) 13 (86.7%) |
0.735 |
|
B symptoms |
|
|
|
|
Absent |
8 (17.0%) |
5 (33.3%) |
0.177 |
|
Present |
39 (83.0%) |
10 (66.7%) |
|
|
HIV Negative Positive |
42 (89.4%) 5 (10.6%) |
14 (93.3%) 1 (6.7%) |
0.651 |
|
First-line chemotherapy regimen R-CHOP Other |
41 (87.2%) 6 (12.8%) |
9 (60.0%) 6 (40.0%) |
0.020 |
|
Second-line chemotherapy regimen |
|
|
|
|
Cytarabine-based |
24 (51.1%) |
8 (53.3%) |
0.878 |
|
Other |
23 (48.9%) |
7 (46.7%) |
|
With a mean follow-up of 32.8 months, median overall survival was 18.6 months (1.3 – 121.0). There were 47 (75.8%) deaths reported. The mean first progression-free survival (PFS) was 16.4 for the sample that did progress, but it was 48.5 months for the first initial 359 patients. Median PFS2 was 7.7 months (0.2–88.6) with 19 (30.6%) second recurrences in the period (Fig. 1 and 2). None of the studied variables correlated with OS or PFS2 in the univariate analysis (Tab. 4, Fig. 3 and 4).
|
Variable |
Categories |
Univariate analysis (OS) |
Univariate analysis (PFS2) |
||
|
n (events) |
p |
n (events) |
p |
||
|
Age |
< 60 years > 60 years |
27 20 |
0.99 |
13 6 |
0.37 |
|
ECOG |
0–1 2–4 |
39 |
0.57 |
17 2 |
0.34 |
|
R-IPI |
Very good Good Poor |
1 16 30 |
0.57 |
2 5 12 |
0.08 |
|
Bulky disease |
Present Absent |
35 12 |
0.58 |
15 4 |
0.89 |
|
Extranodal disease |
Present Absent |
39 8 |
0.44 |
19 0 |
0.03 |
|
B symptoms |
Present Absent |
36 11 |
0.96 |
6 13 |
0.06 |
|
HIV |
Negative Positive |
42 5 |
0.70 |
15 4 |
0.08 |
|
First-line chemotherapy |
R-CHOP Other |
9 38 |
0.76 |
7 12 |
0.41 |
|
Second-line chemotherapy |
Cytarabine-based Other |
25 22 |
0.70 |
8 11 |
0.74 |
|
Second-line RT |
No |
34 13 |
0.30 |
13 6 |
0.08 |
Discussion
Recently, the International Lymphoma Radiation Oncology Group (ILROG) [8] addressed the need for a review and consensus on the use of radiation for relapsed and refractory non-Hodgkin lymphomas. Even though we observe a consistent decrease in the use of radiotherapy for lymphomas [9], patients have been referred to radiation oncology departments particularly after treatment regimens without radiation have failed. In our sample, patients that were not treated with the standard R-CHOP regimen at first, were most suitable to need radiotherapy as part of the salvage treatment (p = 0.02). Our results show a trend to use radiotherapy when usual protocols were not applied.
It is important to stress that radiotherapy has already been compared to regimens containing ASCT in the PARMA trial. Without new evidence for the use of radiation instead of other consolidation regimens it cannot be favored. Nevertheless, this study shows data that raise two important questions. First, the investigation arm for this trial used radiotherapy, even in larger fields, with less technology and lower dose (26 Gy) than is currently used. Second, even amongst highly selected patients for a randomized trial, a fraction of patients received ASCT.
We compared the use of radiotherapy in this setting with other published data. In our sample, 24.2% of relapsed patients did receive radiotherapy. The decrease in the use of radiotherapy is consistent with global trends. It is important to stress that technology has evolved, and patients would have been treated differently from those in the PARMA trial, with smaller volumes and more technology invested into radiation delivery quality assessment [10].
Not every patient that relapses receives ASCT. In the PARMA trial, 41.9% of patients did not respond after high-dose chemotherapy and in the investigation arm, 11% of patients randomized to receive ASCT did not. Our numbers are consistent with this setting, since only four patients did receive ASCT. Even though guidelines stress the importance of ASCT in the relapsed setting, patients usually don’t reach that far. New prospective trials should be designed to investigate the use of radiation as a consolidative treatment after second-line chemotherapy in the setting where ASCT is not possible or feasible.
There is a growing body of evidence in the treatment of recurrent DLBCL with CAR-t cells. This cellular immunotherapy could have a synergy with radiotherapy [11]. As it has been investigated for ASCT, with CAR-T cell treatments the correct order to offer radiotherapy has also been under investigation [12]. This is, nevertheless, a growing application for radiotherapy [13] and new, prospective data is needed.
Radiotherapy for DLBCL has been correlated with toxicities. Even though acute and long-term toxicities are rare [10], particularly with advanced technology, secondary neoplasms are always possible. In the setting of relapsed disease, nevertheless, this is a minor issue compared to the chance of death from the relapsed lymphoma. In special settings, such as patients living with HIV for whom ASCT is not always possible [14], radiotherapy should be more largely used, even with the risk of increased toxicities [15].
Re-irradiation is possible. This study was designed to assess whether patients that receive radiotherapy in the first line regimen would be exposed to radiation again. The chances of in-field relapse for patients treated with radiotherapy in the first line are very small, less than 2% [16]. In our sample, six (9.7%) patients had in-field relapses and were treated with re-irradiation to a dose of 30 Gy, in 2 Gy fractions. No outstanding toxicities were found. Data on re-irradiation for these patients are scarce and our experience may represent a modest but interesting highlight of our study.
Furthermore, there is a paucity of data for patients that have partial response or progress to second-line regimens. Radiotherapy has been investigated before and after ASCT transplantation, with better results when patients receive ASCT with as little residual disease as possible [5]. There is also a place for radiotherapy in the palliative setting. Nevertheless, all those situations haven’t been investigated in prospective trials.
Conclusion
Radiotherapy can be an important tool in the treatment of relapsed DLBCL patients, but its underused. Since the preferred regimen of ASCT regimen is not always feasible, irradiation should be considered mostly in patients that will not undergo ASCT. New trials should be designed to address the role of radiotherapy in the relapse setting as it deserves further studies, mainly in prospective trials.
Ethics approval and consent to participate
Ethics committee authorization was obtained in the local ethics committee according to Brazilian law and the Declaration of Helsinki. All patients have given written consent to participate.
Consent for publication
The author grants the publisher the sole and exclusive license of the full copyright. The authors guarantee that this manuscript has not been previously published elsewhere. The authors declare that any person named as co-author of the contribution is aware of the fact and has agreed to being so named.
Availability of supporting data
data on this research is available on request to the corresponding author
Competing interests
The authors do not have any conflict of interest to declare.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Authors contributions
G.M. was responsible for study design. G.M. was responsible for ethics committee approval. G.M. was responsible for data collection. G.M. and M.N. have written project’s final draft. G.M. was responsible for statistical analysis. H.C. was responsible for overall orientation and manuscript review.
Acknowledgement
We thank Dr. Antonio Brandão, hematologist, for his efforts and help.