Rituximab in the treatment of diffuse large B-cell lymphoma primary of the lung

Abstract

Diffuse large B-cell lymphoma primary of lung (DLBCL-PL) is a rare presentation of extranodal lymphoma, in most cases chemotherapy-based anthracyclines: CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone) has been the treatment, with excellent outcome. The addition of rituximab to CHOP (R-CHOP) has been considered the gold standard in the treatment of nodal DLBCL. Thus, we assess in a large number of cases of DLBCL-PL whether the use of R-CHOP could improve survival in this setting of patients. Forty-two patients with DLBCL-PL, stage IE, age 65 years or younger, were treated with standard R-CHOP, no consolidation radiotherapy or maintenance therapy were considered. They were matched with patients who received CHOP alone to assess efficacy and toxicity. Complete response was observed in 35 patients (83%), and 7 patients were considered failure (16%). The study has a median follow-up of 42.8 months. Actuarial curves at 5 years showed that progression-free survival was 88 % and overall survival was 70 %. The results were not statistically different when compared retrospectively with patients who received CHOP alone. Treatment was well tolerated. The addition of rituximab to chemotherapy did not improve outcome in patients with DLBCL-PL.

Keywords:

• Diffuse large B-cell lymphoma
• Lymphoma primary of lung
• Rituximab
• Doxorubicin
• Chemotherapy

Introduction

Non-Hodgkin's lymphoma (NHL) presents itself as primary lung lesion in <1% of cases and accounts for only 3% of localized extranodal lymphomas.^1–6 Primary lung lymphoma (PLL) was defined as a clonal lymphoid infiltration of lung tissue and always without concurrent involvement of regional lymph nodes or distant viscera. Diffuse large B-cell lymphoma (DLBCL) accounts for the minority of cases of PLL, whereas other lymphoma subtypes, such as malt-associated lymphoid tissue (MALT), are reported at a much higher frequency.^1,2

So far, optimal treatment of DLBCL-PLL has not been established throughout randomized trials. Different treatments have been suggested, including surgery, chemotherapy, radiotherapy, or combined treatments. Until now, the most effective therapy has been the use of anthracycline-containing chemotherapy, as CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone),² with progression-free survival (PFS) and overall survival (OS) of >80% at 5 years. It is well recognized that the addition of rituximab to chemotherapy has led to improved OS and relapse-free survival in nodal lymphomas,⁷ on the basis of these results, we have added rituximab to the standard treatment of PLL. The endpoint of the present study was to determine if the addition of rituximab (R-CHOP) can improve OS in these sets of patients.

Patients and methods

From June 2005 and December 2008, previously untreated 65-year-old patients or younger, no gender difference, diagnosed with PLL and DLBCL histology according to the World Health Organization criteria, with performance status according to the Eastern Cooperative Oncology Group (ECCOG) criteria < , CD 20 + , stage IE, with normal cardiac (measured with left ejection ventricular fraction, normal >50%). Patients who were negative for immunodeficiency virus, hepatitis B and C were included in the study. DLBCL-PLL was either as a single lung lesion, without closely associated lymph nodes’ involvement or multiple lung lesions. Patients with stage III or IV according to the Ann Arbor criteria were excluded for being considered as disseminated disease and have different outcomes. Lung tissues were obtained in all cases by surgery: lobectomy in 3 cases, wedge resection in 32 cases, and lung biopsy in 7 cases. Tissues were sent to laboratory for histochemical study, including immunophenotype with Anti CD20, AntiCD10, Anti CD 5, Anti CD 45, bcl-2, BCL6, and MUM1 according to previous reports.⁸

It was an open-label study that included consecutive patients who fulfill the previous entry criteria. After staging they were treated with an R-CHOP regimen, as follows:

• Cyclophosphamide, 750 mg/m², intravenous (iv), day 1

• Doxorubicin 50 mg/m², iv, day 1

• Vincristine 2 m d, standard dose, iv, day 1.

• Prednisone, 75 mg, standard dose, po, days 1–5

• Rituximab, 375 mg/m², iv, day 1.

All patients received anti-emetic therapy and difenhidramine to prevent acute side effects. Granulocyte colony-stimulating factor was employed only when severe granulocytopenia (<0.5 × 10⁹) was observed to retain the dose-intensity. Each round of treatment was administered every 21 days, if granulocytes and platelets counts were normal. Delay in treatment was considered for hematological toxicity 1 or 2. Patients who developed toxicity grade III and IV were removed from the study.

Before chemotherapy, all patients, including physical examination, completed blood counts, renal and hepatic functional test, serum determination of lactic dehidrogenase and beta 2 microglobulin, determination of immunodeficiency virus, hepatitis B tests. Cardiac and pulmonary function, bone marrow biopsy, and aspirate; computed tomography of thorax, abdomen, and pelvis were performed. After therapy, all patients were carefully restaged, including all positive studies previous to treatment. The International Prognostic Index (IPI) was applicable in all patients. Positron emission tomography was available in our institution in 2007, and only a minor number of cases could be evaluated with the technique.

Efficacy was measured according to International Criteria⁹: complete response (CR) was considered when all clinical and radiological evidence of disease disappear for at least 6 months. Partial response was reduction of 75% of tumor mass. Failure was considered when the response was <50%. The study was approved by the Ethical and Scientific Committee of the Mexican Institute of Social Security (HO-2005-023R).

Statistical analysis

OS and PFS were considered from the day of histology examination using the Kaplan–Meier method. The events were death (including all causes of death) from OS; or relapse for PFS. Age, IPI, tumor size, performance status, and levels of LDH were analyzed by univariate analysis to detect any statistical significance; if any of the mentioned factors showed differences, multivariate analysis according to Cox's regression method was carried out. A P value <0.05 was considered to be statistically significant.

The median follow-up (by December 2010) was 42.8 months (range 23–50 months).

Results

Forty-four patients were considered candidates for the study; two patients refused treatment and were excluded. Thus, 42 patients were considered for analysis. In an intent-to treat analysis all patients were considered evaluate for efficacy and toxicity. They were compared with a historical group of 44 patients that were treated with CHOP alone.

Table 1 shows the main clinical and laboratory characteristics of both groups. Most patients have tumor size between 5 to 10 cm; low clinical risk according to the IPI, good performance status and normal levels of LDH and beta 2 microglobuline. No statistical differences were observed in the two groups.

Table 1. Clinical and laboratory characteristics

	No (%)
	R-CHOP	CHOP
Patients	42 (100)	44 (100)
Age (years)
Range	26–63	29–61
Median	53.4	56.4
Sex
Male	18 (42)	20 (47)
Female	24 (58)	24 (53)
Stage: IE	42 (100)	44 (100)
Tumor size
< 5 cm	4 (9)	2 (4)
5 –10 cm	31 (74)	38 (85)
> 10 cm	7 (16)	4 (8)
International Prognostic Index
Low	39 (91)	42 (95)
Low-intermediate	3 (7)	2 (4)
Lactic dehidrogenase (>2 N)	2 (5)
Beta 2 microglobulin (>2 N)	3 (6)	2 (1)
Smoking history	23 (54)	30 (68)
Performance status
0	12(28)	5 (11)
1	30 (71)	39 (88)

CR was observed in 35 patients (83%) (95 % confidence interval (CI): 78–88%). Five patients relapsed, all with activity outside the lung. Actuarial curves at 5 years showed that PFS was 86 % (95% CI: 80–93 %). Twelve patients died, all secondary to tumor progression. Actuarial curves at 5 years showed that OS was 72% (95%CI: 62–78%). These dates were no different when compared with patients that were treated with standard CHOP (Table 2).

Table 2. Response type and outcome

	Patients (%)
	R-CHOP	CHOP
Complete response	35 (83)	38 (85)
Progression-free survival*	30 (85)	31 (88)
Overall survival*	33 (78)	34 (77)

*Actuarial curves at 5 years.

The treatment was well tolerated, the most frequent hematological toxicity was granulocytopenia grade 1 or 2, delay on treatment was 3–9 days (mean 5.1 days). Dose intensity was not significantly affected. Infection-related granulocytopenia were observed in 12 patients (pneumonias) and were resolved with the use of broad-spectrum antibiotics and G-CSF. Until now, no late toxicities have been observed. Univariate analysis did not shown any statistical differences in the prognostic factors analyzed (data not shown), thus multivariate analysis was not performed.

Discussion

PLL is a rare malignancy of lung, even in tertiary reference centers, as the National Medical Center, that have a coverage population of 31 000 000 and between 2006 and 2010, we diagnosed 3256 cases of malignant lymphoma, and DLBCL PLL accounts for 48 cases (1.4%). For this reason, the absence of controlled clinical trials to define the best treatment in this setting of patients. Moreover, most of the patients with PLL are MALT-type and conservative treatment has been suggested, including no treatment until clinical manifestations of progressive disease. But, DLBCL is a more aggressive disease and needs aggressive treatment even in early stage. The use of surgery has been suggested as an option because it is a localized tumor, but relapse is the rule and chemotherapy is necessary.⁴ Radiotherapy has not been employed as curative attempt, probably for the type of clinical presentation and the risk of lung toxicities. Thus, chemotherapy is the best therapeutic option for DLBCL-PL. CHOP has been demonstrated, which is the gold standard in nodal lymphoma and in most cases of extranodal lymphoma. In n previous paper, we analyzed the clinical presentation and outcome in patients with DLBCL-PLL who were treated with CHOP, in these patients CHOP achieve CR > 80% and PFS and OS >5 years.⁸

The addition of rituximab to standard CHOP has been shown that it increases complete response rate and PFS and OS, especially in younger patients with low and low-intermediate clinical risk.^7,10–12 Thus, the R-CHOP regimen is considered as the gold standard in DLBCL with nodal presentation, even in patients with GCB and non-GCB genotype. Although the prognosis of DLBCL-PLL could be considered good because complete response and overall survival with standard chemotherapy achieve excellent results, we cannot exclude the possibility that rituximab could improve overall survival. For this reason, we considered that the use of R-CHOP will be considered in the treatment of DLBCL-PL and began the present study. We agree that study showed different byass as are low number of patients and that controlled studies are very difficult to develop, taking in consideration that the low number of patients, even in reference centers is very low.

Our results showed that complete response rate, PFS and OS were similar in both groups. No statistical differences were observed, and taking into consideration that it is a uniform population, no prognostic factors could be identified. The use of positron tomography emission (PET) had been shown that improves the image of hematological malignancies and permits defining better treatments. However, our institution is public, with limited economical resources and we did not have available PET for all patients.

It is very difficult to draw definitive conclusions, because all reports in DLBCL-PL are based on retrospective analysis and no controlled clinical trials have been performed, because the number of patients with PLL is very low, even in the tertiary reference center. Rituximab adding to CHOP chemotherapy has been considered the gold standard therapy in DLBCL, with improvement in complete response rate, progression of free survival and overall survival. However, better results have been reported in younger patients and patients with good prognosis features as: IPI low and low-intermediate, without bulky disease and normal levels of lactic dehydrogenase (Pfreuschuh, 2006, Pfreuschuh, 2008).

Thus, the decision of adding rituximab to anthracycline-based chemotherapy in the treatment of PLL in younger patients with good prognosis factors, the population could be considered similar to controlled clinical trials in nodal lymphomas. In our experience, rituximab did not improve the outcome in breast,¹³ gastric,¹⁴ and only in testicular lymphoma rituximab improved overall survival.¹⁵ Thus, it appears that the use of rituximab in extranodal lymphoma remains uncertain. In DLBCL – PLL the use of rituximab appears to be insufficient to increase outcome. We agree that controlled clinical trial is the best study to define this problem, but taking into consideration that to have any statistical difference it will be necessary to recruit at less 188 patients in each arm in a controlled clinical trial, to achieve at less and statistical difference of 10% it is very difficult to perform in this special seeting of patients, even in reference centers. Thus, the use of rituximab will be considered on an individual basis but based on our results we considered that patients with PLL will be treated with standard CHOP, because rituximab did not improve OS.

Conflict of Interest Statement

The authors confirm that the present manuscript has not been published and is it not under consideration from another journal. All authors participate in the conduction of the trial and the final version was revised and authorized for all authors. The present work did not receive any support for external sponsor and was developed with the owner resources of the Mexican Institute of Social Security. All authors disclose any conflict of interest.