A phase III randomized controlled trial of radiation dose optimization in non‐Hodgkin lymphoma‐diffuse large B‐cell lymphoma (DOBL study): Study protocol and design

Abstract

Background

Radiation therapy for diffuse large B‐cell lymphoma (DLBCL) has shown improvement in progression free survival. There is uncertainty about the optimal radiation dose, with heterogeneous doses being used, ranging from 30 to 55 Gy. This trial tests the efficacy of using reduced radiation dose in DLBCL without compromising on long‐term outcomes.

Aims

The primary aim is to assess the noninferiority, as assessed by 2‐year event free survival (EFS), of a dose de‐escalated dose (36 Gy) for DLBCL as compared with a standard dose (45 Gy).

Methods

The Dose Optimization in B cell Lymphomas (DOBL) study is a randomized phase‐III noninferiority trial in a uniform cohort of DLBCL patients receiving immunochemotherapy (R‐CHOP). Patients with stages I to IV of DLBCL eligible for radiotherapy (RT) after completion of at least four cycles of R‐CHOP will be included in the study. Patients will be randomized to standard RT dose of 45 Gy in 25 fractions or reduced dose of 36 Gy in 20 fractions using involved site radiotherapy (ISRT) technique. It is a noninferiority design using a 7% noninferiority margin with a hazard ratio of 1.3, α of 0.05, and β of 0.80. A total of 760 patients will be accrued. Two‐year EFS is the primary outcome measure that will be studied. The experimental arm will be stopped and switched over to control arm if on interim analysis, 25% events occur in the experimental arm.

Discussion

The study is designed to test the noninferiority of radiation dose de‐escalation in a uniform cohort of patients diagnosed with DLBCL and treated with R‐CHOP regimen in the post‐positron emission tomography (PET) era. The trial is rigorously designed and is straightforward to implement. It is statistically powered to answer if reducing the radiation doses does not compromise the clinical outcome in the given patient cohort. This trial will effectively set the standards in radiotherapy doses for DLBCL in the contemporary era.

List of abbreviations

3D‐CRT

three‐dimensional conformal RT

BNLI

British National Lymphoma Investigation

CHOP

Cyclophosphamide, Doxorubicin, Vincristine, and Prednisolone

CI

confidence intervals

CMT

combined modality therapy

CNS

central nervous system

CR

complete response

CT

computed tomography

CTCAE

Common Toxicity Criteria for Adverse Events

CTV

clinical target volume

DLBCL

diffuse large B‐cell lymphoma

ECOG

Eastern Cooperative Oncology Group

EFS

event free survival

EORTC

European Organisation for Research and Treatment of Cancer

FFS

failure free survival

GELA

Groupe d'Etude des Lymphomes de l'Adulte

GTV

gross tumor volume

HR

hazard ratio

IFRT

involved field RT

IMRT

intensity modulated RT

ISRT

involved site RT

LN

lymph node

MLC

multileaf collimators

NHL

non‐Hodgkin lymphoma

ORR

overall response rate

OS

overall survival

PET

positron emission tomography

PFS

progression free survival

PR

partial response

PTV

planning target volume

QOL

quality of life

RT

RT

SAE

serious adverse event

SWOG

Southwest Oncology Group

1. INTRODUCTION

Non‐Hodgkin lymphoma (NHL) is the 12th most common cancer in the world accounting for 2.9% of all cases among men and 2.5% among women. In India, the overall incidence is 2.3% and accounts for 3.3% of cases among males and 1.5% among women.1 Diffuse large B‐cell lymphoma (DLBCL) is the most common subtype. The standard treatment consists of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) chemotherapy with anti‐CD20 antibody rituximab(R). Rituximab has shown to improve event free survival (EFS) and overall survival (OS) in patients with DLBCL.2

The most common site of recurrence after chemotherapy is site of initial disease involvement. Consolidative radiotherapy (RT) decreases the risk of recurrence. Two large cooperative trials from Southwest Oncology Group (SWOG s8736) and Eastern Cooperative Oncology Group (ECOG 1484) have reported improvement in both progression free survival (PFS) and OS in patients with early stage disease with combined modality treatment compared with chemotherapy alone.3, 4 These trials predate the use of rituximab and functional imaging with positron emission tomography (PET) was not employed for response assessment after chemotherapy. However, an updated analysis of the long‐term clinical outcome of SWOG 8736 study showed no difference in the PFS and OS between patients in the CHOP 3 + RT arm vs CHOP 8 arm with continuous treatment failures occurring in both the arms, while the SWOG S0014 study (a phase‐II single arm study) that analyzed the outcome of patients treated with abbreviated course of R‐CHOP (three cycles) combined with RT showed that addition of rituximab did not mitigate the risk of continued relapse.5

In advanced disease, the role of consolidative RT is controversial. Involved field radiotherapy (IFRT) is considered to the sites of bulky disease and partial response (PR). Aviles et al performed a randomized trial comparing CHOP with or without RT to sites of initial bulky disease in patients with DLBCL achieving complete response (CR) to chemotherapy.6 The failure free survival (FFS) and OS were 82% and 87% in those with combined modality therapy vs 55% and 66% with chemotherapy alone. In another trial, Aviles et al also showed improved outcomes in patients with residual disease (defined as tumor mass less than 5 cm) after chemotherapy treated with IFRT.7 The 10‐year Disease Free Survival (DFS) and OS were 86% and 89% compared with patients who did no received RT 32% and 58%, respectively, (P < 0.001).

In the rituximab era, a recent retrospective study from MD Anderson has shown the advantage of RT with R‐CHOP. Patients with DLBCL were treated with at least six cycles of R‐CHOP followed by IFRT. Both OS and PFS improved among patients who received RT. The role of RT was significant in both limited and advanced stage disease. The 5‐year PFS rates for early stage disease treated with RT were 92% versus 73%, respectively, without RT. The 5‐year PFS rates for advance stage disease treated with RT were 76% versus 55%, respectively, without RT. None of the patients experienced treatment failure within the RT field.8

These trials mainly addressed the role of RT in DLBCL. The doses of RT used in these trials have been heterogeneous ranging from 30 to 55 Gy. In the SWOG trial, the dose of RT ranged from 40 to 55 Gy with higher doses for patients with residual disease.3 In the ECOG trial, those achieving a CR received 30 Gy and 40 Gy was given in patients with PR,4 leading to considerable uncertainty about the optimal radiation dose needed in aggressive lymphomas.

Dose response studies for radiation in NHL are sparse with majority of data from old retrospective studies. Fuks and Kaplan in 1973 showed that there was no dose response relation in DLBCL and the recurrence rates were 21% to 37% regardless of the dose of radiation used. The European Organization of Research and Treatment of Cancer (EORTC) RT data from 1970‐1980 showed a 30% local relapse rate at RT dose less than 45 Gy vs 13% in patients who received RT of greater than 45 Gy.9

The data from Princess Margaret Hospital show that DLBCL patients with small volume (less than 2.5 cm) and no improvement in local control was apparent with increasing radiation dose from 20 to 50 Gy, and greater than 90% local control rate was achieved regardless of dose. In patients with medium‐ or large‐bulk disease, defined as 2.5 to 5 cm in size and greater than 5 cm, respectively, an approximately 50% local control rate was achieved with a dose of 20 Gy, rising to 70% at 30 Gy and 80% at 40 Gy with a plateau thereafter and no apparent improvement with additional dose.10 In a retrospective study from British National Lymphoma Investigation (BNLI) by Lamb et al, 11 the local response rate was dependent on radiation dose with 100% response rates for a dose of 45 Gy or more. These studies evaluated the dose of radiation when used alone.

In the current era, all patients with DLBCL are treated with CMT. The more relevant question is the optimal dose of RT for achieving local control and whether dose de‐escalation is feasible in this setting. In a study from British Columbia Cancer Agency, 308 patients with stages I and II DLBCL were treated with CHOP followed by IFRT to doses of 30 to 35 Gy (2 to 3 Gy per fraction). The 10‐year cause‐specific survival rate was 82%. In‐field local failures occurred in 3% of patients.12 In a study from Netherlands, Krol et al evaluated 94 patients with stage I intermediate‐to‐high grade NHL who achieved a CR on computed tomography (CT) imaging after four cycles of CHOP with RT dose 26 versus 40 Gy.13 There was no difference in 5‐year crude in‐field control between patients who received 26 versus 40 Gy (92% vs 95%).

Phan et al treated patients with DLBCL with R‐CHOP (six to eight cycles) with or without RT.8 Overall, 30% had consolidation IFRT following CR to chemotherapy with doses of 30 to 39 Gy. Local control was achieved in 100% of patients with all relapses outside the RT field. Dorth et al evaluated 105 patients of stages I to IV DLBCL treated from 1995‐2009 with consolidation RT after CR to chemotherapy.14 The median RT dose was 30 Gy (range 12‐40 Gy). Majority of patients received R‐CHOP (63%) and underwent functional imaging (89%). The 5‐year in‐field control was 94%. There was no difference in in‐field control among sites that received less than or equal to 20 Gy versus those that received 26 to 30 Gy (94% vs 95%). Local recurrence appeared higher for sites greater than or equal to 10 cm (14% versus 4%, P = 0.06).

There is only one randomized study on dose reduction from the United Kingdom that was carried out by the BNLI group. In this trial by Lowry et al, 1001 patients between 1997 and 2005, 361 with indolent lymphomas, predominantly follicular lymphoma and to a lesser extent marginal zone lymphoma, and 640 with aggressive lymphomas, predominantly DLBCL were randomized between two different radiation schedules: 24 Gy versus 40 to 45 Gy for indolent disease and 30 Gy versus 40 to 45 Gy for aggressive disease.15 At a median follow‐up of 5 years, there was no difference in overall response rate (ORR) between standard and lower‐dose arms, 91% in both arms (P = 0.87). OS and PFS were similar in both arms. In‐field control was similar between the low and high dose groups (82% vs 83%; P = 0.7). However, this study had multiple caveats. This study was not powered to show noninferiority for EFS or PFS, which is considered a standard clinical endpoint in systemic lymphomas. Moreover, the trial included cohort of patients with heterogeneous aggressive histologies with DLBCL being one of them. There is no mention of bulk disease nor was functional imaging (PET/CT) commonly employed to determine objective metabolic response. The study had a mix of patients treated with RT alone or as palliative RT. Moreover, this trial predates the use of rituximab, which is now a standard drug used in the chemotherapy regimen of DLBCL. Additionally, patients accrued in the trial received heterogeneous and aggressive chemotherapy regimens thereby interpreting the role of consolidation RT in the setting of aggressive chemotherapy induction redundant.

In view of these caveats, this trial cannot be considered a reference study for RT dose reduction in the present rituximab and PET era. Hence, there is a requirement for conducting a well‐designed randomized controlled trial to answer the feasibility of RT dose reduction in a uniform cohort of patients diagnosed with DLBCL receiving rituximab + CHOP regimen undergoing response evaluation with PET scan. This is the first randomized trial addressing the issue of optimal radiation dose after chemotherapy in DLBCL.

2. METHODS

2.1. Study aims

The primary aim is to assess the noninferiority, as assessed by EFS, of a dose de‐escalated RT schedule for DLBCL as compared with a standard regimen. The 2‐year EFS is considered as a robust endpoint for determining clinical outcomes in DLBCL patients in the immunochemotherapy era as it has been observed that patients who have a 2‐year EFS have subsequent OS equivalent to that of the age‐ and sex‐matched general population.16 The secondary aim is to evaluate differences in local control, OS, ORR, acute and late toxicity, and quality of life (QOL) of the de‐escalated regimen as compared with the standard.

2.2. Study design

This is an open label, single‐center phase III randomized noninferiority trial. Patients will be randomized after the completion of four to six cycles of chemotherapy just prior to starting radiation. Eligible patients will be randomized in a 1:1 ratio to either reduced dose of 36 Gy in 20 fractions (experimental arm) or standard dose of 45 Gy in 25 fractions (control arm). If the patient is having stable disease or progressive disease after the end of chemotherapy, then the patient will deferred from RT treatment and will be given salvage chemotherapy. Figure 1 shows the trial schema.

Figure 1.

DOBL trial schema

2.3. Informed consent

Informed consent will be obtained by Principal Investigator or personnel designated by the Principal Investigator. Patient will be given appropriate time to decide regarding study participation. Study does not involve minors, pregnant mothers, and neonates. After serving informed consent, patient will be asked to describe the study procedure, benefits, and risks involved to ensure.

2.3.1. Withdrawals

Subjects may withdraw from the study at any point of time.

2.4. Randomization

Randomization will be carried out using stratified permuted block design of varying block sizes. Eligible patients will be randomized in a 1:1 ratio to either reduced dose or standard dose after obtaining informed consent. A computer‐generated list will be created with stratification according response to chemotherapy by PET/CT (CR or PR), bulk of disease (bulky and nonbulky), and revised International Prognostic Index (IPI) (I: very good [0 risk factors]; II: Good (1 or 2 risk factors); III: Poor (3, 4, or 5 risk factors) (Figure 1).17 All randomization will be done at Advanced Centre of Treatment Research and Education in Cancer (ACTREC). Patients will be enrolled and assigned to interventions by principal investigator or the person designated by the principal investigator. In order to have adequate participant enrolment to reach target sample size, there will be uniform accrual of patients over a period of 4 years. Every year, the trial will be monitored for the accrual rate by the external monitoring committee, and based on their recommendations, strategies will be put in place to ascertain that the target sample size is reached within the given time period.

2.5. Research setting

The study is being conducted at Tata Memorial Hospital (TMH), Mumbai and ACTREC Navi Mumbai.

2.6. Participants and research eligibility

2.6.1. Inclusion criteria

Patients aged 18 to 65 years with ECOG 0‐3 and histological diagnosis of NHL‐DLBCL stages I to IV eligible for RT after R‐CHOP chemotherapy will be included in the study. Patients should receive at least four cycles of R‐CHOP chemotherapy. Patients with bulky disease (greater than or equal to 7 cm as measured longest measurement by CT scan) and partial response (PR) according to The Lugano Classification after four cycles of R‐CHOP will be included in the study.18 Patients should be able to understand and willing to provide informed consent for participation in the trial. Informed consent will be taken by the principal investigator and coprincipal investigators of the trial.

2.6.2. Exclusion criteria

Patients with HIV positive status, relapse or progression of disease during chemotherapy, systemic DLBCL with central nervous system (CNS) involvement, primary extranodal lymphomas (ie, testicular, stomach, and intestine), patients with greater than three extranodal sites, and those with double expresser/double hit/triple hit and primary CNS lymphoma will be excluded from the study. Based on the Cheson's criteria, patients with Progressive disease(PD) or patients with Stable Disease(SD) will be excluded from the study. Patients with a prior history of chemotherapy or RT for lymphomas will also be excluded.

2.7. Study interventions

2.7.1. Chemotherapy

All patients will receive four to six cycles of R‐CHOP chemotherapy based on the stage of the disease. This consists of Rituximab, Cyclophosphamide, Doxorubicin, Vincristine, and Prednisolone. Patients with stages I and II will receive abbreviated chemotherapy four cycles followed by radiation or six cycles followed by radiation depending on response and bulk of disease. In stages III and IV, patients will receive six cycles of chemotherapy. Patients with a high intermediate/high IPI score, particularly reflecting the presence of a high level of lactate dehydrogenase and involvement of more than one extranodal site are at much higher risk for CNS relapse and intrathecal prophylaxis with methotrexate will be incorporated into the systemic regimen. Radiation will be considered sequential to chemotherapy.

2.7.2. Radiotherapy

Patient preparation and treatment simulation

These procedures will be followed in all patients irrespective of arm allocation. The patient positioning and immobilization would vary depending on the site being treated. In general, patient will be positioned supine with an immobilization device. Orthogonal lasers will be used for patient alignment. Laser marks made on the thermoplastic mask will help in setting up the patient during treatment.

• For head and neck lymphomas: Patient immobilization will be achieved by a four‐clamp thermoplastic mask in supine position with arms by the side of the body.

• For axilla: Immobilization will be achieved in akimbo position.

• For mediastinal tumors: Patient will be positioned supine with knee rest with arms in overhead position. A four‐clamp thermoplastic mask will be used for immobilization.

• Abdomen and pelvis: Patient will be positioned supine with knee rest with arms over chest. A four‐clamp thermoplastic mask will be used for immobilization.

A planning contrast enhanced CT scan will be acquired in actual treatment position with a slice thickness of 2.5 to 5.0 mm depending on the region of interest. Staging/Diagnostic 18^F fluorodeoxy glucose PET/CT (¹⁹F‐FDG PET/CT) will be fused with the post treatment planning CT scans for delineation of target volumes for involved site RT (ISRT) technique.

Contouring

The following target volumes will be delineated on the planning CT and would be included as part of the clinical target volume (CTV). Gross tumor volume (GTV) will be classified as (a) prechemotherapy GTV: GTV as outlined based on imaging abnormalities suggestive of gross lymphomatous involvement before chemotherapy. (b) Postchemotherapy GTV: GTV includes the residual disease if any after chemotherapy. CTV includes the nodal volumes delineated, which would include the lymph node (LN) region that was originally involved modified for normal tissue boundaries. The lateral extent would be limited to post chemotherapy volume in case of mediastinum and paraaortic nodes. The target volumes will be in concordance to the article by Illidge et al.15 All LN regions are irradiated simultaneously as a single RT series. The initially involved LN regions are to be irradiated together in a single field as far as anatomically possible.

Supradiaphragmatic LN regions:

• 1Unilateral cervical involvement: RT of the ipsilateral cervical + supraclavicular LN. CTV would extend from base of skull to the clavicle including the ipsilateral level Ib, II, III, and IV.
• 2Bilateral cervical involvement: RT of the neck including all cervical + supraclavicular LN. CTV will be defined bilaterally as defined for unilateral neck involvement including submental LN.
• 3Axillary nodal involvement: CTV to include axillary, supraclavicular, and infraclavicular nodes.
• 4Mediastinum: CTV includes mediastinal nodes, bilateral hilar, and bilateral supraclavicular nodes. For lateral extent, postchemotherapy volume to be considered.

Infradiaphragmatic regions:

• 5Para‐aortic: CTV includes the para‐aortic nodal region from T10‐T11 down to L5‐S1 (with at least 2‐cm margin to prechemotherapy volume craniocaudally). Laterally, the involved LN is included with a minimum margin of 1.5 cm, at least encompassing the transversus process of the lumbar vertebrae.
• 6Unilateral/bilateral iliac: CTV includes the iliac/pelvic LN from L4‐L5 to the bottom of obturator foramen.
• 7Groin: CTV includes femoral and inguinal LNs.

Planning target volume (PTV) includes the entire CTV plus a minimum 3D margin of 5 mm. The organs at risk would vary depending on the site of treatment. Head and neck: eyeball, optic nerve, optic chiasm, lens, pituitary, temporal lobes, brainstem, spinal cord, parotids, oral cavity, thyroid, larynx, and esophagus; axilla and mediastinum: lungs, heart, spinal cord, esophagus; and abdomen and inguinal: bowel loops, spinal cord, kidneys, liver, duodenum, and femoral heads.

Radiation planning

All patients will be treated with three‐dimensional conformal RT (3D‐CRT) or intensity modulated RT (IMRT). In case of mediastinum and extranodal sites, combination of multiple beams would be required. Multileaf collimators (MLC) will be used for shaping the fields. Seven‐millimeter margin for MLC will be used while conforming to the PTV. The minimum PTV dose (to a volume of at least 0.03 cc) must not fall below 95% of the prescription dose, and the maximum dose within the PTV must not exceed 107% of the prescribed dose.

Dose prescription and radiation schedule

Patients will receive treatment 5 days per week, in a once‐daily fraction, 1.8 Gy per fraction. The total dose in control arm is 45 Gy in 25 fractions over 5 weeks and in experimental arm 36 Gy in 20 fractions over 4 weeks. RT commences after recovery of blood parameters, especially of leukocytes and thrombocytes, 4 weeks if possible but at most 6 weeks after the end of chemotherapy. Electronic portal images will be obtained during the course of treatment on the first day of treatment and once a week thereafter and will be compared with the digitally reconstructed images from CT scan. The maximum number of allowable treatment interruptions would be seven working days (excluding holidays and cumulative).

Relevant concomitant care and interventions that are permitted or prohibited during the trial

During the trial, the participants are permitted to receive drugs for their comorbid conditions (diabetes, hypertension, etc) and premedications that will be required during the chemotherapy and medications for managing the acute radiation side effects. The participants are prohibited to take any drugs/medicines to which they are allergic.

2.8. Assessment during treatment

All patients will be required to complete QOL assessment (FACT‐G and FACT‐Lym) prior to initiating chemotherapy and again prior to radiation. After four cycles of chemotherapy, FDG‐PET‐CT scan will be done for response assessment. During radiation therapy, all patients will be assessed weekly by a radiation oncologist, and acute toxicity will be documented using National Cancer Institute‐common toxicity criteria for adverse events version 4.0 (CTCAE V.4.0). At treatment conclusion, a repeat evaluation of QOL will be done (Table 1).

Table 1.

Assessments in DOBL study

Required Investigations	Prestudy	Pre‐RT	Post‐RT at 3 to 4 mo	Six Monthly Up to 2 y	Greater than 2 y—Yearly
Physical
Case history	X	X	X	X	X
Clinical examination	X	X	X	X	X
Hematology
CBC with differential and ESR	X	X
TFT	X			X a	X a
Biochemistry
RFT, LFT	X	X		X a	X a
S. LDH	X	X		X a	X a
Radiology
Chest X‐ray	X
FDG‐PET/CT	X	X	X a
Other investigations
Cardiac—ECG, 2D Echo	X	X		X c
PFT	X			X c
Adverse events
CTC version 4	X	X b	X	X	X
QOL assessments	X	X b	X	X d

Abbreviations: CBC, Complete Blood Count; CTC, Common  Toxicity Criteria; CT, computed tomography; ECG, Electro Cardio Graph; ESR, Erythrocyte Sedimentation Rate; FDG‐PET/CT: FluoroDeoxy Glucose‐Positron Emission Tomography/Computerized Tomography; LDH, Lactate Dehydrogenase; LFT, Liver Fuction Test; PET, positron emission tomography; QOL, Quality of life; RFT, Renal Fuction test; RT, radiotherapy; TFT, Thyroid function Test.

^aAs clinically indicated.

^bAt RT conclusion as well.

^cAt 1 y.

^dAt 1 and 2 y.

2.9. Follow‐up

All patients will be followed up as follows: first visit at 3 ± 1 month and then six monthly—from completion of RT to 2 years. Thereafter, the patient will followed up yearly. At each visit, clinical examination, weight, and performance status, includes the CTCAE V.4.0 scale for acute reactions and late effects. QOL (FACT‐G and FACT‐Lym) will be done at 3 months after completion of RT followed by at 1 and 2 years. A FDG‐PET‐CT scan at 3 months after treatment is repeated if there was PR after completion of chemotherapy (with PET positive disease, Deauville criteria 3, and above) and in the event clinical suspicion of relapse. To promote participant and complete follow‐up, the patients will be contacted telephonically to attend the dedicated trial clinic (Table 1).

2.10. Measurable endpoints

The primary endpoint is 2‐year EFS. The secondary endpoints are 2‐year local control rates, 2‐year OS, ORR, acute and late toxicity, and QOL scores.

2.11. Statistical considerations

Using a group sequential design, the study with one interim and one final analysis, 369 events are required to prove noninferiority that translates to a hazard ratio of 1.3 (experimental/control). Given the above conditions, 840 evaluable patients will be required to be accrued uniformly over 4 years with an additional follow‐up of 2 years. The total duration of the study is expected to be 6 years from the time the first patient is entered.

The projected numbers of EFS failure events and the nominal significance levels for rejecting the null or alternate hypothesis at interim analyses, along with the projected timing and accrual, are shown in Table 2. At the planned interim analysis, the one‐sided P value from the log‐rank test assessing treatment efficacy with respect to EFS will be compared with the nominal significance levels in the above table. These levels are based on the LanDeMets alpha spending function and O'Brien‐Fleming boundary. If the computed P value for efficacy is less than or equal to the nominal significance level boundary for rejecting the H0 (efficacy), then accrual to the trial will be stopped (if applicable).

Table 2.

The nominal significance levels for rejecting the null or alternate hypothesis at interim analyses

Analysis		Reject H0	H0−	H0+
Interim	186	0.006	2.538	3.47
Final	372	0.048	1.662	3.58

The sample size is calculated for primary end point (EFS). The expected EFS in control arm is 70% at 2 years based on average EFS of patients treated with R‐CHOP in trials from the United States and Groupe d'Etude des Lymphomes de l'Adulte (GELA) group. The maximum acceptable noninferiority margin is 7% with a hazard ratio of 1.3 corresponding to 63% EFS in experimental arm at 2 years. With a type I error (α) of 0.05 (one sided) and power of test 80%, expected duration of recruitment of 4 years, minimal follow‐up of 2 years, and the total number of events needed 369, the sample size is 754 patients (both arms). Considering an attrition rate of 10%, the total sample size is estimated to be 840 patients (420 in each arm). One interim analysis is planned for the primary endpoint at 2.5 years when 25% of the likely numbers of events (92 events) have been observed. Using the O'Brien stopping rule, the boundaries at interim analysis and final analysis are 3.75 and 1.64 respectively.

2.11.1. Stopping rule criteria

The experimental arm (reduced dose) will be stopped and switched over to control arm if on interim analysis, more than 25% events occur.

2.12. Data collection

All trial data will be maintained by the principal investigator of the study at ACTREC, TMC, Mumbai, India. The patient data will be anonymized and annotated to maintain the confidentiality of the participants. The access to final data set will be with principal and coprincipal investigators.

2.12.1. Treatment data

All data will to be collected and documented on standardized case report forms. The chemotherapy cycles, overall treatment time, breaks if any in radiation treatment will all be captured.

2.12.2. Toxicity evaluation

Toxicity will be assessed using CTCAE V.4.0 (Supplementary material S1). CTCAE forms will be filled before starting radiation, weekly during radiation treatment, and on each scheduled follow‐up. The acute toxicities during both chemotherapy and radiation therapy and in the follow‐up period will be managed according to the standard institutional protocol.

2.12.3. Quality of life

FACT‐G and FACT‐Lym (Supplementary material S2) will be used for evaluating QOL of patients. All patients will undergo QOL evaluation before chemotherapy, before radiation, at radiation conclusion, at 3 to 4 months, at 1 year, and then at 2 years post radiation.

2.12.4. Clinical outcome data

In addition, the study will also record the status of disease at each follow‐up. A record of each examinations finding will be kept. Systemic work up will be performed only once a year, and status of regional and distant control will be reported accordingly.

2.12.5. Protocol compliance

Inability to receive entire planned chemotherapy schedule will be considered as major violation. Inability to achieve PTV coverage with 95% isodose will be considered as minor violation.

2.13. Reporting adverse events

Adverse events will be recorded on the CTCAE V.4.0 during treatment and follow‐up. All serious adverse events (SAE) are reported to the institutional review board within seven working days and within 24 hours in the case of fatal or life‐threatening events. Toxicity arising out of systemic chemotherapy prior to randomization will not be considered as trial related injury or a related SAE since this is the standard of care in the management of DLBCL.

2.14. Trial monitoring

The trial will be monitored at regular interval by the institutional data and safety monitoring board, and its report will be submitted to the ethics committee and institutional review board. An independent external data and safety monitoring board has been constituted to independently monitor the trial at regular intervals. All protocol amendments and modifications in the informed consent forms will be duly conveyed to the institutional review board and ethics committee.

2.15. Data analysis plan

EFS will be calculated from the time of randomization till relapse or progression or death due to any cause or death due to disease. OS will be calculated from the time of randomization to death due to any cause/death due to disease. To estimate EFS and OS, Kaplan Meier curves will be presented for the two arms, with corresponding median EFS and OS, and survival estimates at 2 years with corresponding 90% confidence intervals (CI). The noninferiority of the experimental arm will be established based on both intention to treat and per protocol analysis.

Cox proportional hazard analysis will be used for assessing the effect of different variables on the survival specifically the stratification factors. The treatment effect in terms of hazard ratio (HR) will be presented corresponding to 90% CI. Decision making will be based on the upper limit of the 90% CI around the HR for EFS. If the upper limit is below 1.34, the lower dose RT arm will be declared noninferior to the standard arm. Survival will also be compared between the two arms by log rank test.

Local control and response rates will be compared between the two arms by the logistic regression adjusting for stratification factors. The proportion of patients achieving CR rates will be compared by logistic regression adjusting for stratification factors. The incidence of toxicity between the arms will be compared by logistic regression or chi‐square test. As the trial is assessing noninferiority, interpretation will be based on the boundaries of the confidence intervals.

2.15.1. FACT‐G and FACT‐Lym

All patients who have completed FACT‐G and FACT‐Lymphoma questionnaire for baseline and at least one follow‐up time point will be considered for analysis. Analysis of covariance, using the baseline FACT‐G and FACT‐Lymphoma score as a covariate will be used to determine if there is a difference in overall and lymphoma‐specific QOL between the treatment arms. The global QOL score over the time interval considered will be compared between two arms using Area under Curve (AUC) statistic for the ease of interpretation and reporting. Generalized estimating equations (GEE) modeling will also be used as a subsidiary analysis for reporting the longitudinal changes in QOL data.19

2.16. Ethical considerations

The study has been approved by Tata Memorial Centre Ethics review committee, and it poses no obvious risks to subjects. All the data will be saved in a locked area and can be only accessible by authorized personnel.

3. DISCUSSION

There is growing body of evidence that shows combined modality therapy for NHL‐DLBCL is associated with increase in OS and PFS.4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 However, heterogeneous radiation doses are reported in the studies, and most of the studies are retrospective in nature. There is only one prospective study conducted by BNLI group addressing the issue of RT dose de‐escalation in lymphomas. This study included patients of diagnosed with indolent and aggressive lymphomas. As mentioned above, this study had a major limitation as patients received heterogeneous and aggressive chemotherapy regimens thereby reducing the effectiveness of consolidation RT. Moreover, this study was powered to study local control as a primary endpoint.15 However, lymphomas are considered systemic diseases; hence, a more appropriate and relevant endpoint would be EFS, and therefore, we wanted to study if maintaining local control by de‐escalating radiation doses eventually translates to optimal EFS. Hence, the present study was designed to study the 2‐year EFS as a primary outcome measure between the two arms.

In the post‐rituximab and PET era, this is the first randomized study that addresses the issue of radiation doses reduction in NHL‐DLBCL This study will also help to know whether reduced dose of radiation, ie, 36 Gy will be sufficient for patients with partial responders after six cycles of chemotherapy.

With the evolution of radiation delivery techniques, efforts are being made to not only reduce radiation doses in lymphomas but also the radiation volumes. In the contemporary era, modern RT techniques such as ISRT (as advocated by the International Lymphoma Study Group)20 and involved node RT (INRT as advocated by the EORTC group).21 The present randomized trial will utilize these contemporary RT techniques especially the ISRT technique for both the arms.

4. CONCLUSION

As we are transitioning from IFRT to ISRT in lymphomas in the post‐rituximab and post‐PET era, the present study is designed to test the noninferiority of radiation dose de‐escalation in the combined modality treatment of NHL DLBCL in the rituximab era using R‐CHOP chemotherapy. This trial will effectively set the standards in RT doses in DLBCL in the contemporary era.

TRIAL REGISTRATION

clinicaltrials.gov identifier: NCT02964858. Date: November 12, 2016.

DATA SHARING POLICY

Readers can access extra data (individual patient data) through principal investigator JSG at godajayantsastri@gmail.com. The data will not be shared with study sponsors who do not have any role in study design, collection, management, analysis, interpretation, and writing of data. The final authority for the data sharing and aforementioned tasks lies with the study principal investigator.

CONFLICT OF INTEREST

None of the authors have any competing interests.

AUTHORS' CONTRIBUTION

All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, J.S.G.; Methodology, J.S.G., S.C.L., S.K.; Investigation, J.S.G., S.L.N.K., H.J., B.B., S.E.; Formal Analysis, J.S.G., S.C.L., S.K.; Resources, J.S.G.; Writing ‐ Review & Editing, J.S.G., B.B., H.J.; Visualization, J.S.G., S.C.L.; Supervision, J.S.G., H.J., B.B.; Funding Acquisition, J.S.G.

Supporting information

Data S1: Supporting information

Goda JS, Lewis SC, Laskar S, et al. A phase III randomized controlled trial of radiation dose optimization in non‐Hodgkin lymphoma‐diffuse large B‐cell lymphoma (DOBL study): Study protocol and design. Cancer Reports. 2019;2:e1161. 10.1002/cnr2.1161