Treatment strategies for nodal and gastrointestinal follicular lymphoma: Current status and future development

Abstract

In recent years, therapies for follicular lymphoma (FL) have steadily improved. A series of phase III trials comparing the effect of rituximab with chemotherapy vs chemotherapy alone in treating FL have indicated significant improvements in progression-free survival (PFS) and overall survival. Recent studies have found that prolonged response durations and PFS were obtained with maintenance therapy using rituximab or interferon after completion of first line therapy. For patients with relapsed or refractory FL, phase II studies have assessed the effectiveness of combination therapies using a Toll-like receptor-9 agonist (1018ISS), oblimersen sodium (a Bcl-2 antisense oligonucleotide), bendamustine, and rituximab, as well as veltuzumab, a new humanized anti-CD20 antibody, and epratuzumab. In addition, the effectiveness of yttrium-90 ibritumomab tiuxetan and iodine-131 tositumomab as radioimmunotherapies has been reported. Furthermore, three phase III studies on an idiotype vaccine are near completion. Unfortunately, these vaccines, which appeared highly effective in phase I and II trials, do not appear to result in prolonged PFS. This report will summarize the current knowledge on therapies for treatment of FL, and will conclude with a brief discussion of feasible future options for effective treatments. Lastly, we added descriptions of the management of gastrointestinal FL, which is considered to be controversial because it is rare.

Key Words: Anti-CD20 monoclonal antibody (rituximab), Follicular lymphoma, Idiotype vaccines, Immunoradiotherapy, Treatment strategies

INTRODUCTION

Indolent lymphoma is a lymphoma that tends to grow and spread slowly, in contrast to an aggressive lymphoma which grows and spreads quickly (Definition of Indolent lymphoma. MedTerms™: the medical reference for Medicine Net.com. http://www.medterms.com/script/main/hp.asp). While the majority of indolent lymphomas, which make up 30%-35% of all non-Hodgkin’s lymphoma (NHL) subtypes, are follicular lymphoma (FL)[1], their incidence is low in Western, Asian, and less-developed countries[2]. The incidence of FL accounts for only about 6.7% of all malignant lymphomas in Japan[3]. Its incidence is rapidly increasing in Western and Asian countries[4].

FLs are often detected by the swelling of systemic lymph nodes in adults. FLs are pathologically characterized by medium-sized centrocytes and large-sized centroblasts found in normal follicles as proliferative cells. Most FLs are detected at an advanced stage (Ann Arbor stage III or IV), and it was considered difficult to find a better treatment than the “watch and wait strategy” during the asymptomatic phase[5-8]. In recent years, various novel treatment options have been developed for patients with advanced stage FL, including (1) monoclonal antibodies (rituximab, veltuzumab (a new humanized anti-CD20 antibody), and epratuzumab) with conventional chemotherapy; (2) radioimmunotherapy (yttrium-90 ibritumomab tiuxetan and iodine-131 tositumomab); and (3) idiotype vaccines. These therapeutic procedures have a major impact on overall survival (OS) of patients.

Hiddemann et al[9] published a review article on the treatment of FL in 2005, however, in recent years, a review article especially on the recent progress in therapeutic procedures for FL has not been published. In this paper, current knowledge regarding therapies for treating FL will be summarized. By reviewing recent publications (in particular, those published in the past 2 years), a variety of treatment options, including novel monoclonal antibodies, immunoradiotherapy, combination therapies of these with conventional chemotherapies, and idiotype vaccines will be discussed. Idiotype vaccines are a nascent treatment alternative for which phase III studies are now near completion[10]. If successful, these vaccines can potentially be developed as customized anti-cancer treatments optimized for each patient.

The gastrointestinal (GI) tract is the most common site of extra-nodal NHL, accounting for 30%-40% of primary extra-nodal NHL[11,12]. The histologic subtypes in most primary GI-NHL are mucosal-associated lymphoid tissue (MALT) lymphoma, or high-grade and aggressive B-cell lymphoma[13-16]. GI-FL is rare, and the frequency of this entity accounts for around 2% of GI-NHL[16-18]. However, recently, because capsule endoscopy and double balloon endoscopy of the small intestine were generalized widely in Japan, the number of primary GI-FL patients has tended to increase. In this review manuscript, I describe the treatment strategies for primary GI-FL, which remains controversial.

TREATMENT IS GUIDED BY THE HISTOLOGIC SUBTYPE AND EXTENT OF DISEASE

At present, the following factors are taken into account when deciding the course of treatment for FL: (1) histopathological types and grade of malignancy; (2) lesions where FL developed (in particular, those of extra-nodal origin); and (3) staging classification (Ann-Arbor classification).

Histopathological types and grade of malignancy

In FL, pathological subtypes are classified as grade 1-3a and 3b depending on the number of centroblasts per field of view. FL grade 3b, which is based on the presence or absence of residual centrocytes (3a vs 3b), was optional in the 2001 WHO classification[4], but is now mandatory[19]. Details of the grade of malignancy are shown below: grade 1: Number of centroblasts is 0 to 5 per high-power histological view; grade 2: Number of centroblasts is 6 to 15 per high-power histological view; grade 3: Number of centroblasts is more than 15 per high-power histological view; grade 3a: Centrocytes are present; grade 3b: Centroblasts proliferate in sheet formation and no centrocytes are present.

In nodal FL, several studies suggest that this histological grading is a good predictor of prognosis[20,21]. However, the treatment is not decided directly by this histological grading alone, and is decided mainly by staging (extent of disease) or both staging and histological grading[22].

In nodal FL, the proportions of grade 1, grade 2, and grade 3 are 40%-60%, 25%-35%, and 20%, respectively[23], while those of grade 1, grade 2 and grade 3 in GI-FL are 84.4%, 11.3%, and 4.3%, respectively[24]. The proportion of grade 1 in GI-FL accounts for about 85% and commands a majority compared with that in nodal FL. Furthermore, on staging, the proportions of stage I and II are 66.3% and 26.9%, respectively, and that of stage I plus II (early stage) is 93.2%. The degrees of grading are considered to be similar to those of staging, which is to say that in early-stage FL, the patients at stage I and II, and with grade 1 and 2 (Grade 1 and 2 FL is histologically subclassified as “Low-grade” FL[22]) command a majority. With regard to treatment strategies, especially in nodal FL, radiation therapy will be selected first.

In recent years, even if FL patients were found to be in the early stages (stage I or II), rituximab was included as a treatment strategy in those with nodal or extra-nodal FL to prolong survival, in fact, rather than the so-called “Watch and Wait strategy”, aggressive therapies including mainly rituximab tend to be started in the earlier stages in Japan[25].

Lastly, in GI-FL, because the disease lesions are limited, several types of therapeutic options, for instance, surgical resections (plus adjuvant chemotherapy with rituximab, or rituximab alone), or in cases with no symptoms, chemotherapy plus rituximab or the “Watch and Wait strategy” are selected. There is no standard regimen, and the treatment policy is controversial in GI-FL[24]. Conversely, it has been reported that in nodal FL, most cases are found to be in stage III or IV at the diagnosis with FL[22], however, the proportions of grade 1 and 2 are about 50% and 30%, respectively (the proportion of grade 1 plus 2 is 80%)[23], and the degree of grading is considered to be dissimilar to that of staging. The number of patients with stage III or IV and low-risk or low-grade (grade 1 or 2) FL seems to be comparatively high. There is no standard therapy for advanced, but low-grade FL to date[24], however, a combination of classical chemotherapy and rituximab is now considered to be a main therapy for advanced FL, because it has been reported that this combination prolonged survival compared with several classical chemotherapies alone.

The treatments for nodal FL and GI-FL are summarized as follows: Most cases with GI-FL have been found to have focal disease and an early-stage condition at diagnosis, with a histological grading of low-grade, while nodal FL is almost always found at an advanced stage. However, the degrees of cellular malignancies were considered to be divided into two groups of low-grade and high-grade, and the proportions were reported to be about 80% and 20%, respectively. When physicians discuss the treatment strategy for nodal FL and GI-FL, they should consider the differences in the status between these two groups, however, both the treatment regimens for stage III-IV low-grade FL, and stage III-IV high-grade FL do not differ at present.

Furthermore, there is little or no difference in the natural clinical course or the response to treatment between the two groups with histological grade 1 and 2 FL. There is a consensus that physicians initiating different therapeutic options for FL seems meaningless[22].

Biological malignancy is also an important factor which is directly associated with the degree of differentiation of tumor cells, the increased speed of tumor cells themselves, infiltrative growth into the surrounding tissue, and liability to metastasize, which is important for prognosis. With this in mind, the following subjects should be discussed: (1) How many degrees, and how to add the factors with differences in histological grading of the tumor cells themselves, to the factors regarding the selection of therapeutic procedures induced by the differences in staging; and (2) Is the present method of grading FL cells really suitable for FL? This is one of the problems which is very difficult to solve, but very important when discussing the treatment strategies for FL in the future.

Grade 3b FL shows the same clinical features as diffuse, large B-cell lymphoma (malignant lymphoma with a medium grade of malignancy). This grade does not show a slow and indolent course, but rather the same clinical course as an aggressive lymphoma. Some studies have identified biological differences between these 2 subtypes, with most instances of FL Grade 3b being more closely related to diffuse large B-cell lymphoma at the molecular level[26,27]. Consequently, the treatment policy for Grade 3b should be guided in a similar fashion to that of aggressive lymphoma. Given that tumor progression is very different, Grade 3b FL needs to be distinguished from the other grades of FL.

Clinical staging of FL

Clinical staging of nodal FLs is determined using the Ann-Arbor Clinical Staging system[28], while GI-FLs are classified by the Lugano staging classification of GI tract lymphoma[29], which is the modified version of the Ann-Arbor staging system for GI tract lymphomas.

Most indolent lymphomas like FL are discovered at an advanced stage because they grow slowly and only a small number of FL cases have systemic symptoms.

Most FLs are detected at an advanced stage (Ann Arbor stage III or IV), and it was considered difficult to find a better treatment than the “Watch and Wait strategy” during the asymptomatic phase[5-7].

However, randomized controlled trials (RCT) using a new anti-CD20 antibody have shown that combination therapy with traditional chemotherapy plus rituximab prolonged OS in patients compared with conventional chemotherapy alone. Rather than “Watch and Wait”, combination therapy with monoclonal antibody and chemotherapy is considered standard therapy even when non-symptomatic FL is discovered. Even if FL patients are found to be in the early stages (stage I or II), rituximab has been introduced as a treatment strategy in patients with nodal FL to prolong survival, in fact, aggressive therapies including mainly rituximab have recently been initiated in the earlier stages in Japan.

Radiation therapy can treat cases in the limited stage (stage IA). Cases need to be assessed to determine whether they are in stage I using a PET-CT scan (difficult to detect FL), systemic CT scan, bone marrow puncture, or aspiration.

For FL cases in the limited stage (Ann-Arbor staging IA), additional radiation therapy directed at focal lesions after 3 or 4 courses of chemotherapy can be considered in addition to the integral therapy.

In cases where a portion of the tumor remains, additional radiation therapy can be performed after chemotherapy.

Chemotherapy should be performed for cases at a stage ≥ II.

The Lugano staging classification of GI tract lymphoma[29] is shown in Table 1.

Table 1 Lugano staging classification of gastrointestinal tract lymphoma[29].

Stage
I	Tumor confined to GI tract
	Stage primary site or multiple, noncontiguous lesions
II	Tumor extending into abdomen from primary GI site
	Nodal involvement
II1	Local (paragastric in cases of gastric lymphoma and paraintestinal for intestinal lymphoma)
II2	Distant (mesenteric in the case of an intestinal primary; otherwise paraaortic, paracaval, pelvic, inguinal)
IIE	Penetration of serosa to involve adjacent organs or tissues [enumerate actual site of involvement, e.g. IIE (pancreas), IIE (large intestine), IIE (post intestinal wall)]
	Where there is both nodal involvement and penetration involving adjacent organs, the stage is denoted using both a subscript (1 or 2) and E, e.g. II1E (pancreas)
IV	Disseminated extra-nodal involvement, or a GI tract lesion with supradiaphragmatic nodal involvement

GI: Gastrointestinal.

This is the modified version of the Ann Arbor staging system for GI tract lymphomas. Yamamoto et al[24] reported that 128 (66.3%) and 52 (26.9%) of 193 GI-FL cases were at stage I and II, respectively. Surprisingly, over 90% of all GI-FL cases which have been reported in the English literature were at the early stages, and had localized disease. Yamamoto et al[24] suggested that this was because of the low tendency of GI-FL to disseminate outside the GI tract, and the higher frequency of grade 1 lymphoma in GI-FL cases. Furthermore, the GI-FL cases, in which sufficient examinations and successful treatments were performed due to non-advanced stages and no symptoms, were likely to be reported and submitted for publication in the English literature. The possibility of bias in the selection of the reported cases or case series should be raised. Nodal involvement can be found in GI-FL cases, and this status is classified as stage II₁ (involvement of local lymph nodes) or II₂ (involvement of distant lymph nodes).

Disease distribution (unifocal or multifocal) is important for selecting treatments in GI-FL patients.

In recent years, Yamamoto et al[24], Kodama et al[30] and Higuchi et al[31] have claimed that other disease lesions in the small intestine were newly and secondarily found in over 70% of GI-FL patients who had disease lesions in the GI tract except the small intestine in which capsule endoscopy and double-balloon endoscopy were performed. If the patients were found to be multifocal, rituximab with or without conventional chemotherapy may be administered.

THERAPIES FOR NODAL FOLLICULAR LYMPHOMAS

Traditional therapies for FL

FL cases are categorized into pathological grades (1 to 3) based on the proportion of cells comprised by centroblasts. This grading system is an important factor in determining prognosis. For example, grade 3 lymphoma is the most malignant and has the fastest progression, however, chemotherapy can be an effective treatment. In contrast, grade 1 represents the least malignant lymphoma and has a very slow progression where the patient may die 10 years after the time of diagnosis. However, this lymphoma is unaffected by chemotherapy. Thus, prognosis between grades 1 and 3 is strikingly different[32-35].

In cases of NHL not restricted to FL, and in those where GI invasion is observed, patients risk lethal complications if the chemotherapy is too strong. These complications include bleeding or GI perforations due to tumor necrosis. Thus, GI complications must be considered when selecting the appropriate treatment regimen. However, choosing mild chemotherapy or low-intensity radiation therapy due to GI lesions may also result in insufficient treatment of systemic lymphoma. Therefore, the success or failure of a treatment depends on both the grade and degree of disease spread.

Several therapies have been tested for treating GI-FL. In grade 1 and 2 FLs, some treatment options include an alkylating agent alone, a combination therapy with multiple drugs, and high-dose chemotherapy in combination with hematopoietic stem cell transplantation. However, no more than 10% of patients maintain complete remission during the 5-year post-treatment period. Additionally, while the median survival time is seven to 10 years, the survival rate declines progressively and becomes lower than that of aggressive lymphoma after approximately 15 years[36]. Given that the effects of chemotherapy have so far been suboptimal, and disease progression is very gradual even without treatment, some have suggested initiating treatment only after symptoms actually emerge. In the interim, they recommend careful observation. In the early stages (I and II) of FL, radiation therapy has been effective, resulting in a control rate of 95% and a 10-year progression-free survival (PFS) rate of 40%-52%. In these circumstances, radiation therapy is considered standard treatment because there is promise of healing or remission for an extensive time period. However, the likelihood of radiation therapy actually being performed clinically is low, even in the US[37]. While studies are being conducted to observe the effects of rituximab and irradiation combination therapy, no long-term observational data exist. Recently, therapeutic drug-targeting molecules such as rituximab have played an important role in prolonging the lives of FL patients. In the treatment of systemic FL, 6 cycles of CHOP and rituximab (R-CHOP) combination therapy appear to be the most promising. During a phase III RCT, R-CHOP was found to significantly prolong survival compared to CHOP therapy alone[38,39].

Randomized control studies of rituximab plus conventional chemotherapy vs conventional chemotherapy alone

Results of RCTs comparing response rate, duration, disease-free survival (DFS), and OS between patients treated with conventional chemotherapy and those treated with rituximab plus conventional chemotherapy have been reported (Tables 2 and 3). Details are described below.

Table 2 Results of different randomized trials comparing chemotherapy with chemotherapy plus rituximab in previously untreated patients with follicular lymphoma.

Study	No. of patients, pathological types, staging	Median age (yr)	Complete response rate	End point
				Median progression-free survival	Event-free survival	Median response duration	Time to treatment failure	Overall survival
CVP[40]	n = 321, FL, Stage II-IV	52	P < 0.0001b			P < 0.0001b	P < 0.001b	P = 0.029a
R-CVP	n = 162	Unknown	41%	34 mo		Prolonged than CVP alone (lower relapse rate than CVP alone)	Prolonged than CVP alone	83% (at 4 yr)
CVP	n = 159	Unknown	11%	15 mo				77% (at 4 yr)
GLSG[38,39]	n = 428, FL, Stage III or IV(advanced stage)	55	P > 0.05 (not significant)			P = 0.001b	P < 0.001b	P = 0.016a
R-CHOP	n = 223	54	20%	Not reached		Lower relapse rate than CHOP alone	28/233 (12.0%)	95% (at 2 yr)
CHOP	n = 205	57	17%	31 mo			61/205 (29.8%)	90% (at 2 yr)
FL2000[41]	n = 358, FL, 89% > stage II	61	P = 0.001b		P = 0.001b	P = 0.012a		P > 0.05 (not significant)
R-CHVP/IFN	n = 175	Unknown	67% (at 18 mo)	Not reached	53% (at 5 yr)	64% (at 4 yr)		84% (at 5 yr)
CHVP/IFN	n = 183	Unknown	50% (at 18 mo)	35 mo	37% (at 5 yr)	44% (at 4 yr)		79% (at 5 yr)
OSHO[42]	n = 201, FL, Stage II and IV	59	P = 0.0004b	P < 0.0001b	P < 0.0001b	P < 0.0001b		P = 0.0096b
R-MCP	n = 105	60	52/105 (50%)	Not reached	Not reached	Not reached		87% (at 4 yr)
MCP	n = 96	57	25/96 (25%)	28.8 mo	26 mo	35 mo		74% (at 4 yr)

^aP < 0.05,

^bP < 0.01, statistically significant. (R-)CVP: (Rituximab plus) cyclophosphamide, vincristine, and prednisone; FL: Follicular lymphoma; GLSG: German Low-Grade Lymphoma Study Group; (R-)CHOP: (Rituximab plus) cyclophosphamide, doxorubicin, vincristine, and prednisone; (R-)CHVP/IFN: (Rituximab plus) cyclophosphamide, adriamycin, etoposide, prednisolone plus interferon-α2a; (R-)MCP: (Rituximab plus) mitoxantrone, chlorambucil, and prednisolone; OSHO: East German Study Group Hematology and Oncology.

Table 3 Results of different randomized trials comparing chemotherapy with chemotherapy plus rituximab in relapsed or refractory follicular lymphomas.

Study	No. of patients, target disease	Median age (yr)	Complete response rate	End point
				Median progression-free survival	Overall survival
GLSG[43]	n = 65, relapsed FL		P = 0.011a	P = 0.0139a	P = 0.0943 (not significant)
R-FCM	n = 35	60	94%	Near 36 mo	90% (at 2 yr)
FCM	n = 30	59.5	70%	21 mo	70% (at 2 yr)
EORTC, HOVON, NCIC, CTG, BNLI, ALLG, NLG and EORTC Data Center-multicenter study[44]	n = 465, relapsed/resistant FL		P < 0.001b	P < 0.001b	P = 0.096 (not significant)
R-CHOP	n = 234	54	29.5%	33.1 mo	82.5% (at 3 yr)
CHOP	n = 231	55	15.6%	20.2 mo	71.9% (at 3 yr)

^aP < 0.05,

^bP < 0.01, statistically significant. GLSG: German Low-Grade Lymphoma Study Group; FL: Follicular lymphoma; EORTC: European Organization for Research and Treatment of Cancer; HOVON: Hemato-Oncologie voor Volwassenen Nederland; NCIC: National Cancer Institute of Canada; CTG: Clinical Trial Group (Canada); BNLI: British National Lymphoma Investigation; ALLG: Australasian Leukaemia and Lymphoma Group; NLG: Nordic Lymphoma Group; (R-)CHOP: (Rituximab plus) cyclophosphamide, doxorubicin, vincristine, and prednisone; (R-)FCM: (Rituximab plus) fludarabine, cyclophosphamide, mitoxantrone.

Rituximab plus chemotherapy trials for previously untreated FL: Several studies have shown that conventional chemotherapy plus rituximab for previously untreated FL prolonged various types of survival (Table 2).

A multicenter randomized phase III trial of 321 patients with previously untreated FL compared cyclophosphamide, vincristine, prednisolone (CVP) alone to rituximab plus CVP (R-CVP, n = 162)[40]. OS was higher with R-CVP compared to CVP alone (OS, P = 0.029; 4-year OS: 83% vs 77%, respectively).

The randomized trial conducted by the German Low-grade Lymphoma Study Group involved treating patients with previously untreated FL with 6-8 cycles of CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) combined with rituximab (R-CHOP, n = 223) or CHOP alone (n = 205). A significantly higher overall response (OR) rate (96% vs 90%, respectively, P = 0.011) and prolonged duration of remission (P = 0.001) were achieved with R-CHOP. The first 3-year OS with 6 deaths in the R-CHOP group was superior to that of 17 deaths in the CHOP group (P = 0.016)[38,39].

The FL-2000 randomized trial involved subjecting patients with previously untreated FL to treatment with either CHVP (cyclophosphamide, adriamycin, etoposide and prednisolone) plus interferon-α2a (CHVP+I arm) or CHVP combined with rituximab plus interferon (R-CHVP+I arm). After a median follow-up of 5 years, event-free survival was significantly different between the two arms (37% in the CHVP+I arm vs 53% in the R-CHVP+I arm, P = 0.001). However, the 5-year OS rate between the two arms of CHVP+I and R-CHVP+I was not significantly different[41].

Herold et al[42] randomly assigned previously untreated patients with advanced MALT lymphoma, FL, or mantle cell lymphoma (MCL) to one of two groups, either mitoxantrone, chlorambucil, prednisolone plus rituximab (R-MCP, n = 181) or MCP alone (n = 177). All patients who achieved a complete or partial remission were treated with maintenance interferon until they relapsed. They reported the results from the primary analysis population of patients with FL alone, who constituted the majority of patients (56%) recruited to the trial (n = 201; R-MCP, n = 105; MCP, n = 96). OS rate in the R-MCP group significantly improved compared to the MCP alone group (4-year OS rate, 87% vs 74%, respectively, P = 0.0096).

Randomized trials of rituximab with chemotherapy in relapsed or refractory FL: A randomized trial which involved fludarabine, cyclophosphamide, mitoxantrone (FCM) alone or rituximab plus FCM (R-FCM)[43] was conducted with 30 FL patients allocated to the FCM and 35 FL patients to the R-FCM treatment groups (total 65 patients). The OR rate [complete response (CR) plus partial response rate] for R-FCM was 94% compared to 70% for FCM alone. The R-FCM arm was significantly superior with respect to PFS (P = 0.0139) compared to FCM for FL[43] (Table 3).

In another randomized trial, a total of 465 patients with relapsed or resistant FL were randomly assigned to CHOP or R-CHOP induction therapy groups. Those in complete remission or partial remission were then randomly assigned to maintenance therapy with rituximab or an observation alone group. Median PFS for rituximab maintenance or observation alone after the second randomization was significantly different (51.5 mo vs 14.9 mo, respectively). In relapsed or resistant FL, rituximab maintenance improves PFS after R-CHOP induction[44] (Table 3).

Maintenance therapy with rituximab or interferon after first-line therapy

The question as to whether or not maintenance therapy should be carried out has been raised. In one case involving rituximab, Hainsworth et al[45] concluded that maintenance therapy is not required because first-line therapy with rituximab was effective in cases without maintenance therapy. Conversely, several studies have shown that rituximab as a maintenance therapy after conventional chemotherapy or interferon as maintenance therapy after conventional chemotherapy plus rituximab prolonged response duration and PFS (Table 4).

Table 4 Results of different randomized trials comparing no maintenance with maintenance therapy with rituximab or interferon after first line therapy in patients with follicular lymphoma.

Author(s) or studies, maintenance therapy	No. of patients, target disease and randomization to therapies	Median age (yr)	Initial therapy,n (%)	CR rate	End point
					Median PFS	Median response duration	Overall survival
GLSG[43,46]	n = 105, relapsed or refractory FL					P = 0.035a
R-FCM	n = 52	59	R-FCM: 41 (79)		Not reached	32/41 (78%)
			FCM: 11 (21)
Observation	n = 53	61	R-FCM: 40 (75)		Not reached	21/40 (53%)
			FCM: 13 (25)
ECOG1496 study[47]	n = 311, stage III-IV indolent lymphoma (282 of 311 had FL)	61	CVP	P = 0.00006b		P = 0.012a	All: P = 0.05
							FL: P = 0.08 (trend towards significance)
R-CVP	n = 153		CVP	22%		All: 68% (at 3 yr)	All: 92% (at 3 yr)
						FL: 64% (at 3 yr)	FL: 91% (at 3 yr)
Observation	n = 158		CVP	7%		All: 33% (at 3 yr)	All: 86% (at 3 yr)
						FL: 33% (at 3 yr)	FL: 86% (at 3 yr)

^aP < 0.05,

^bP < 0.01, statistically significant. CR: Complete response; PFS: Progression-free survival; FL: Follicular lymphoma; GLSG: German Low-Grade Lymphoma Study Group; (R-)FCM: (Rituximab plus) fludarabine, cyclophosphamide, mitoxantrone; ECOG: Eastern Cooperative Oncology Group; (R-)CVP: (Rituximab plus) cyclophosphamide, vincristine, and prednisone.

Forstpointner et al[43,46] reported that maintenance therapy with rituximab was effective after salvage therapy with rituximab along with conventional chemotherapy, and significantly prolonged the response duration in patients with FL. Hochster et al[47] also reported the results of the randomized phase III ECOG1496 trial, and found that maintenance rituximab after cyclophosphamide, vincristine, and prednisone prolonged PFS in patients with stage III-IV (advanced) untreated indolent lymphoma (282 out of a total of 311 patients had FL). Among patients with FL alone, OS at 3 years in the patients with and without maintenance rituximab was 91% and 86%, respectively; HR = 0.6, log-rank one-sided P = 0.08). Further, Herold et al[42] reported that rituximab added to first-line MCP, followed by interferon maintenance, prolonged survival in patients with advanced FL.

New anti-cancer agents with rituximab in patients with relapsed or refractory FL

Several studies have recently reported on phase II trials involving combination therapy with new anticancer agents plus rituximab for relapsed or refractory FL patients. Friedberg et al[48] conducted a Phase II trial of a new anti-cancer agent, Toll-like receptor-9 (TLR-9) agonist, combined with rituximab in patients with relapsed and refractory FL. TLR-9 agonists have pleiotropic effects on both innate and adaptive immune systems, including increased antigen expression, enhanced antibody-dependent cell-mediated cytotoxicity, and T helper cell type 1 shift in the immune response. They reported that clinical responses were observed in 48% of patients and that the overall median PFS was 9 mo. Pro et al[49] conducted a Phase II multicenter study of combination therapy with oblimersen sodium, a Bcl-2 antisense oligonucleotide, and rituximab in recurrent B-cell NHL patients. They reported an OR rate in relapsed FL patients of 60% (eight complete remissions and four partial remissions out of 20 patients), and that the combination of oblimersen sodium and rituximab seemed to be most effective in patients with indolent NHL including FL. Further, Robinson et al[50] evaluated the phase II multicenter study of bendamustine plus rituximab in patients with relapsed indolent B-cell and MCL who showed no resistance to prior rituximab. They reported that OR rate, median duration of response, and median PFS time were 92% (41% CR), 21 and 23 mo, respectively, and concluded that the combination of bendamustine plus rituximab is suitable for relapsed indolent lymphoma centering on FL and MCL.

New anti-CD20 antibody for indolent B-cell lymphoma

Veltuzumab is a humanized anti-CD20 antibody with structure-function differences from chimeric rituximab. Morschhauser et al[51] evaluated a multicenter phase I/II dose-finding trial in relapsed/refractory B-cell NHL evaluating the efficacy of veltuzumab. They reported that in FL, 24 (44%) of 55 patients had objective responses, and CR were seen in 15 (27%) patients. In marginal zone lymphoma, 5 (83%) of 6 patients had OR. Veltuzumab appeared to be safe and active at all doses tested with rituximab, including dose levels less than those typically used, encouraging further study.

New antibodies against targets other than CD20

Epratuzumab, a humanized IgG1 unconjugated anti-CD22 antibody: Leonard et al[52] conducted an international multicenter, phase 2 trial in patients with recurrent FL and reported durable CR with a combination of epratuzumab and rituximab. They reported an OR rate of 85% in patients with FL who had Follicular Lymphoma International Prognostic Index (FLIPI) risk scores of 0 or 1 (n = 13), whereas 28 patients with intermediate or high-risk FLIPI scores (≥ 2) had an OR rate of 39%.

Inotuzumab ozogamicin (CMC-544), the calicheamicin-conjugated anti-CD22 monoclonal antibody and rituximab: Takeshita et al[53] studied the effect of CMC-544, a calicheamicin-conjugated anti-CD22 monoclonal antibody, used alone or in combination with rituximab, and investigated the quantitative alternations on the target molecules of CD20, CD22, CD55and CD59 in cells obtained from patients with B-cell malignancies or Dauji and Raji cells. The reduction in CD55 and the preservation of CD20 after incubation with CMC-544 support the rationale for the combined use of CMC-544 and rituximab[48]. This new antibody may also be used for the treatment of FL in the future.

Galiximab, a human-primate chimeric anti-CD80 antibody: Galiximab is a human-primate chimeric anti-CD80 antibody with excellent tolerability and single-agent effectiveness for recurrent FL[54]. A phase I/II trial of the combination of rituximab plus galiximab demonstrated a 64% response rate and a 12.1 mo median PFS[55]. A phase III trial comparing rituximab with the combination of rituximab plus galiximab in patients with relapsed FL is ongoing.

Other potential new antibodies against targets other than CD20, CD22, or CD80: Phase I/II trials testing new CD52 (alemtuzumab), CD2 [MEDI-507 (siplizumab)], CD30 [SGN-30 and MDX-060 (iratumumab)], and CD40 (SGN-40)[56] antibodies are now ongoing or have been completed in aggressive lymphoma patients. It is possible that these new antibodies will be used for the treatment of FL in the near future.

Radioimmunotherapy for FL

In 2005, Kanzius et al[57,58] provided new systems and methods for radio-frequency-induced hyperthermia plus radioimmunotherapy using ⁹⁰Yttrium-ibritumomab tiuxetan or ¹³¹Iodine-tositumomab for patients with NHL containing FL.

Recently, radioimmunotherapy using ⁹⁰Yttrium-ibritumomab tiuxetan or ¹³¹Iodine-tositumomab, which are produced from anti-CD20 monoclonal antibodies combined with radioactive materials, has been performed for refractory or relapsed FL patients. Recent reports in this field are listed below:

¹³¹Iodine-rituximab for relapsed indolent B-cell lymphoma (FL and MALT lymphoma): Illidge et al[59] evaluated a novel protocol using 4 weekly infusions of 375 mg/m² rituximab followed by 2 fractions of ¹³¹Iodine-rituximab, preceded by a 100 mg/m² predose of rituximab in patients with relapsed indolent B-cell lymphoma (FL and MALT lymphoma). OR rate was 94%, with a CR rate of 50%. Median time to progression was 20 mo, which is significantly longer than that for the last qualifying chemotherapy (P = 0.001).

⁹⁰Yttrium-ibritumomab tiuxetan for previously untreated FL: Jacobs et al[60] reported on a trial involving patients with previously untreated FL who received CHOP-R (three cycles of CHOP and four additional weekly rituximab cycles). CR rate after CHOP-R, as assessed by CT and PET imaging, was 40% and 46%, respectively. CR rate further improved following ⁹⁰Yttrium-ibritumomab tiuxetan) to 82% and 89%, respectively.

⁹⁰Yttrium-ibritumomab tiuxetan for advanced FL after first remission: Morschhauser et al[61] described that consolidation of first remission with ⁹⁰Yttrium-ibritumomab tiuxetan in advanced-stage FL is highly effective with no unexpected toxicities, prolonging PFS by 2 years and resulting in high partial remission-to-complete remission conversion rates regardless of the type of first-line induction treatment.

⁹⁰Yttrium-ibritumomab tiuxetan for relapsed or refractory indolent B-cell lymphoma (containing FL): Tobinai et al[62] reported that radioimmunotherapy with ⁹⁰Yttrium-ibritumomab tiuxetan is safe and highly effective in patients with relapsed or refractory indolent B-cell lymphoma (small lymphocytic, lymphoplasmacytic, mantle cell, follicular, MALT, and splenic marginal zone lymphomas) pretreated with rituximab-based therapy. They also reported OR and CR rates of 83% and 68%, respectively, and a median PFS of 9.6 mo.

Idiotype vaccine

Idiotype vaccines are based on active immunization of patients against their own tumor idiotype, and could lead to an even more effective and sustained anti-tumor response than the passive infusion of anti-idiotype antibodies[10]. Vaccine therapies, which utilize antibodies against the characteristic tumor idiotype for each FL patient, are expected to result in anti-tumor effects, improved response rates, and prolonged survival times. These benefits are anticipated because these vaccine therapies can potentially be developed as personalized anti-cancer treatments optimized for each patient. Several phase I/II clinical trials for idiotype vaccines in FL have been conducted in the last 20 years, of which about 10 have been published and demonstrate the effectiveness of idiotype vaccines[63-71]. In particular, Inogès et al[69] reported that most vaccinated patients with FL had significantly longer DFS after second remission. These encouraging results from phase I/II trials initiated three phase III RCTs (Biovest, Genitope, and Favrille). These phase III trials are now essentially completed and have been evaluated (Table 5). While there initially seemed to be a significant prolongation of PFS during the early stages of the trials, all of these phase III trials ultimately failed to prolong PFS[10].

Table 5 Results of phase III clinical trials of idiotype vaccines in follicular lymphoma.

Study	No. of patients, target disease	Regimen of induction therapy	Eligibility	Vaccine type	Vaccine production methodology	Primary end point
						Survival types, P-value, significance
Biovest	n = 177, untreated FL	PACE or R-CHOP	CR	Id-KLH/GM-CSF	Rescue hybridoma	Disease-free survival, P > 0.05, not significant
Genitope	n = 287, untreated FL	CVP	CR, PR	Id-KLH/GM-CSF	Recombinant DNA	Progression-free survival, P > 0.05, not significant
Favrille	n = 349, untreated or recurrent FL	Rituximab	CR, PR, SD	Id-KLH/GM-CSF	Recombinant DNA	Time to progression, P > 0.05, not significant

FL: Follicular lymphoma; PACE: Prednisolone, doxorubicin, cyclophosphamide and etoposide; R-CHOP: Rituximab plus cyclophosphamide, doxorubicin, vincristine, prednisolone; CVP: Cyclophosphamide, vincristine, prednisolone; CR: Complete response; PR: Partial response; SD: Stable disease; Id-KLH: Idiotype-keyhole limpet hemocyanin; GM-CSF: Granulocyte-macrophage colony stimulating factor.

McCormick et al[72] recently produced patient specific recombinant idiotype vaccines in plants against follicular B cell lymphoma and conducted a phase I trial. They reported that more than 70% of patients developed cellular or humoral immune responses, and that 47% of patients developed antigen-specific responses[72]. Results from future phase II/III trials will be of great interest.

THERAPIES FOR GASTROINTESTINAL FOLLICULAR LYMPHOMAS

Very recently, Yamamoto et al[24] performed a literature review of Japanese cases with GI-FL. This review on GI-FL is well written and I believe may help with regard to the recent confusion surrounding the selection of treatment for GI-FL patients.

I would like to summarize this editorial, especially the treatment of GI-FL patients, by focusing on the main points covered in this review article.

Introduction of GI-FL

The GI tract is the most common site of extra-nodal NHL, accounting for 30%-40% of all primary extra-nodal NHL[11,12]. While GI involvement in aggressive NHL (nodal aggressive NHL with involvement of the GI tract) is common, primary GI-NHL is rare, and GI-FL is very rare. It has been reported that the histologic subtypes in most primary GI-NHL are MALT lymphoma or high-grade B-cell lymphoma[13-16]. GI-FL is rare, and this histology represents around 2% of all GI-NHL[16-18]. In addition, over 90% of GI-FL cases which have been published in the English literature to date were early stages (stage I and II)[24]. Very recently, Yamamoto et al[24] performed a review of the literature on patients with GI-FL, and claimed that the diagnosis, prognosis, and treatment of GI-FL remain largely unknown mainly because of its rare frequency. However, recently, because capsule endoscopy and double balloon endoscopy of the small intestine were generalized in Japan, the number of primary GI-FL patients has tended to increase.

Prognosis in patients with GI-FL

The long-term clinical outcome of GI-FL is as yet unclear. Damaj et al[73] reported that there was no significant difference between two groups of patients treated by either conventional chemotherapy with rituximab, and patients without therapy based on the “Watch and Wait” concept regarding the median time to disease progression (37.5 mo) and the median relapse-free time (31 mo).

Treatment for GI-FL

Many patients with GI-FL have been identified in the early stages, however, the disease lesion is not unifocal, but multifocal, with another lesion in the small intestine, and rituximab with or without chemotherapy for many non-advanced GI-FL cases may be initiated according to the regimens for nodal FL[24].

However, because it has been reported that there was little difference in survival between GI-FL patients with and without therapies, and GI-FL itself is very rare, the management of GI-FL remains controversial at present[24].

In recent years as capsule endoscopy and double balloon endoscopy of the small intestine have been generalized in Japan, we have the opportunity to identify GI-FLs, particularly in patients with asymptomatic GI-FL. In addition, the therapeutic options in these patients such as whether conventional chemotherapy with rituximab should be initiated or no therapy with the “Watch and Wait” concept requires careful consideration. This decision may be very important.

Yamamoto et al[24] recently performed a literature review and reported the actual numbers in each therapeutic category. Observation without therapy was only about 25%, while some kind of therapy was performed in 75% of all primary GI-FL patients. Chemotherapy (or combination with other therapies) was performed in about 50% of all primary GI-FL patients.

Chemotherapy or other therapies tend to be performed more frequently in patients with GI-FL rather than observation without therapy based on the “Watch and Wait” concept. The reasons why GI-FL patients undergo chemotherapy are as follows: (1) because of increasing acceptability, generalization, dissemination, and recognition of the concepts of primary GI-FL and the chemotherapy regimens as treatment strategies for GI-FL; and (2) because it has been reported that the combination of conventional chemotherapy with rituximab can prolong the survival of nodal FL patients compared with conventional chemotherapy alone, rituximab-containing regimens have been administered to a growing number of GI-FL patients. Future prospective studies on the treatment of GI-FL patients are awaited with great interest.

CURRENT AND FUTURE DEVELOPMENTS

Therapeutic monoclonal antibodies, such as the anti-CD20 antibody, provide significant benefits to patients with NHL. In particular, since 2000 various phase III trials have utilized RCTs to compare the treatment effects of conventional chemotherapies vs the addition of rituximab to these therapies. The results indicated that the combination of rituximab and conventional chemotherapy was generally superior both in terms of response rates and survival times. Thus, this combination therapy is gradually becoming the standard. Recently, phase I and II trial results assessing the initial use of a novel anti-CD2 antibody, alone or in combination with rituximab, in patients with recurrent aggressive lymphoma have been published. Determining the usefulness of these new antibodies compared to rituximab requires randomized comparative trials or a demonstration of their effectiveness in rituximab-refractory or relapsed patients after completion of first-line rituximab therapy.

Phase III trials comparing galiximab with rituximab combined with galiximab are in progress. Similar studies will be necessary in the future to assess the effectiveness of many novel anti-CD20 antibodies. Some anticipate that rituximab, in combination with these new anti-CD20 antibodies, will improve response rates and prolong PFS.

Radioimmunotherapy with ⁹⁰Yttrium-ibritumomab tiuxetan or ¹³¹Iodine-rituximab is safe and highly effective in patients with relapsed or refractory (resistant to rituximab-based therapy) FL. The effectiveness of these radiolabeled antibodies needs to be compared to other kinds of therapies in randomized trials.

Idiotype vaccines are immunotherapeutic products which have been developed to induce active and long-lasting immune responses against lymphoma cells. Most of these vaccines use the tumor B cell idiotype (the variable region of the surface immunoglobulin) as a tumor-specific antigen. Several phase I/II clinical trials of idiotype vaccines in FL have been conducted over the last 20 years, of which about 10 have been published and demonstrate the effectiveness of these vaccines. These encouraging results, which seem to indicate that the vaccines prolonged PFS, initiated three phase III RCTs (Biovest, Genitope, and Favrille), however, all of these phase III trials ultimately failed to prolong PFS.

Primary GI-FL is very rare, and this histology represents around 2% of all GI-NHL. Many patients with GI-FL have been found at the early stages, however, the disease lesion is not unifocal, but multifocal, with another lesion in the small intestine, and rituximab with or without chemotherapy for many non-advanced GI-FL cases may also be initiated based on the regimens for nodal FL. However, it has been reported that there was little difference in survival between GI-FL patients with and without therapies. Furthermore, GI-FL itself is very rare, and the management of GI-FL remains controversial at present. Yamamoto et al[24] recently performed a literature review and reported that only about 25% of primary GI-FL patients underwent observation without therapy, and chemotherapy (or combination with other therapies) was performed in about 50% of all primary GI-FL patients. Ongoing research on biomarkers to guide individualized GI-FL therapy may provide invaluable information which will lead to the establishment of a standard therapeutic regimen.

CONCLUSION

This article summarized the existing information on FL treatment through an extensive literature review which included the latest research on drug therapies. The various treatments reviewed included novel monoclonal antibodies, immunoradiotherapy, and the combination of these therapies with conventional chemotherapy. Moreover, idiotype vaccines, a nascent treatment alternative for which phase III trials are now near completion, were discussed. If successful, the latter can potentially be developed as personalized anti-cancer treatments optimized for each patient. This article concluded with a brief discussion on potential future developments in the treatment of FL.