Survival after progression in patients with follicular lymphoma: analysis of prognostic factors

S. Montoto1, A. López-Guillermo1,+, A. Ferrer1, M. Camós1, A. Álvarez-Larrán1, F. Bosch1, J. Bladé1, F. Cervantes1, J. Esteve1, F. Cobo1, D. Colomer2, E. Campo2 and E. Montserrat1

1Department of Hematology and 2Hematopathology Unit, Institut d’Investigacions Biomèdiques August Pi i Sunyer, Hospital Clínic, Barcelona, Spain

Received 6 August 2001; revised 6 November 2001; accepted 19 November 2001.


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background

The purpose of this study was to identify prognostic parameters for patients with follicular lymphoma (FL) in first progression/relapse. These would be useful for selection of high-risk patients for inclusion in trials aimed at determining the effect of new treatment approaches in such patients.

Patients and methods

Ninety patients (48 male, 42 female, median age 56 years) diagnosed with FL, in a single institution during a 20 year period and relapsing/progressing after an initial response to therapy, were recruited. The main end-point of the study was survival from progression (SFP). Univariate and multivariate analyses were performed, including among the predictive variables the response duration (RD) after the initial treatment and the main features of the patients at the first progression or relapse.

Results

Five-year SFP was 47% (95% confidence interval 35% to 58%). Patients with RD following initial therapy >2 years had a longer SFP (5-year SFP 63 versus 33%, P = 0.012). Other variables with prognostic interest for SFP were stage at diagnosis and the following variables at relapse: age, bulky disease, performance status, serum lactate dehydrogenase level, serum ß2-microglobulin level, bone marrow involvement, stage and International Prognostic Index rating. In the multivariate analysis, poor performance status at progression and a RD <2 years were the most important unfavorable variables to predict SFP.

Conclusion

In patients with FL, RD along with performance status at progression are features that predict SFP. These variables could thus be useful to select candidates for experimental treatments.

Key words: follicular lymphoma, outcome, progression


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The majority of patients with follicular lymphoma (FL) present with disseminated disease and have a relatively indolent course, the median survival being 6–10 years. However, FL is not curable with conventional therapies, and most patients die from causes related to the disease, frequently after having experienced many relapses [14]. Given the above facts, new potentially more effective, but also potentially more toxic, treatments are increasingly being used to treat FL. These include purine analogs, either alone or in combination [5, 6], monoclonal antibodies [79], biological agents and stem cell transplantation [1016]. Although there is evidence that these approaches can be useful, their exact role and the best timing for such treatments are not yet well established. On the other hand, although the toxicity related to some of these approaches appears to be acceptable [7, 17], the complications caused by other strategies are of concern. For example, the incidence of myelodysplastic syndrome/acute myeloid leukemia (MDS/AML) after stem cell transplantation has been found to be 4–12% [14, 15, 18]. Taking all these facts into account, as well as the responsiveness of the disease to standard treatments and its rather indolent nature, experimental treatments are usually reserved for patients in relapse. Therefore, the analysis of prognostic factors at relapse should be useful to select patients for whom inclusion in experimental trials is warranted.

The aim of the present study was to analyze the prognostic features in a series of patients with FL from a single institution, focusing on the variables at the time of the first relapse or progression that predict survival.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Initial characteristics and treatment
Two-hundred-and-one patients (median age 54 years; 107 males, 94 females) consecutively diagnosed with FL between January 1977 and December 1997 in a single institution were the basis for this study. The histological distribution according to the Revised European-American Classification of Lymphoid Neoplasms/World Health Organization (REAL/WHO) classification [19, 20] was as follows: grade I, 85 patients (42%); grade II, 97 patients (49%); grade III, nine patients (4%); unclassified, 10 patients (5%). Twenty patients (10%) presented with stage I; 19 (9%) with stage II; 31 (15%) with stage III; and 131 (66%) with stage IV. Extranodal involvement was demonstrated in 141 patients, including bone marrow in 123 cases (61%). The distribution of the patients according to the International Prognostic Index (IPI) [21] was as follows: low risk, 89 patients (49%); low/intermediate risk, 50 patients (27%); high/intermediate risk, 24 patients (13%); and high risk, 20 patients (11%).

Initial treatment consisted of: monotherapy with alkylating agents, 29 patients (14%); COP chemotherapy (cyclophosphamide, vincristine and prednisone), 28 patients (14%); adriamycin-containing regimens (in most cases CHOP: cyclophosphamide, adriamycin, vincristine and prednisone), 126 patients (63%); and other treatments, 18 patients.

Among 191 assessable patients, 87 (46%) attained a complete response (CR), 77 (40%) a partial response (PR), while 27 patients (14%) were considered non-responders. After a median follow-up of 4.5 years, 41 of the 87 CR patients (47%) and 49 of 77 PR patients (64%) had progressed. Overall, the median response duration (RD) was 2.3 years. These 90 patients (48 male, 42 female; median age 56 years) who responded to treatment and eventually relapsed/progressed were the subject of the present study. The main characteristics at diagnosis and relapse are listed in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Main characteristics of 90 patients with follicular lymphoma at diagnosis and at relapse
 
Staging, evaluation at response and follow-up procedures
Initial staging procedures included computed tomography of the thorax, abdomen and pelvis, as well as a bone marrow biopsy. Post-therapy restaging included the repetition of the previously abnormal tests and/or biopsies.

CR was defined as the disappearance of tumor masses and disease-related symptoms, as well as the normalization of the initially abnormal tests and/or biopsies, lasting for at least 1 month. PR was considered as measurable lesions decreased by at least 50% [22]. Patients not included in these categories and those who died before completing treatment were considered non-responders.

The follow-up surveillance procedures after treatment consisted of physical examination, blood count tests and biochemistry, and chest X-ray (if initially abnormal) every 3 months during the first year, every 4 months during the second year, every 6 months during the next 3 years and every year afterwards. When the abdomen was the principally involved site, abdominal computer tomography (CT) scans were performed every 6 months during the first year and yearly for the following 5 years. Molecular follow-up was performed only during the last 4 years (1996–2000).

At the time of clinical relapse or progression, a new lymph node biopsy was performed whenever possible. However, such a procedure was carried out in only half of the patients due to older age, poor performance status or patient refusal. For the same reasons, a new bone marrow biopsy could only be obtained in 42 patients. In addition, restaging of the patients included CT scans of the thorax, abdomen and pelvis.

Parameters evaluated
In each patient the following data were recorded and evaluated for prognosis. (i) Initial variables: age (<60 versus >=60 years), gender, performance status [according to the Eastern Cooperative Oncology Group (ECOG) scale], B-symptoms, histological subtype, hemoglobin, leukocyte count, lymphocyte count, leukemic expression, platelet count, serum albumin, serum lactate dehydrogenase (LDH) and serum ß2-microglobulin (ß2M) levels, number of nodal and extranodal involved sites, presence of splenomegaly, Ann Arbor stage, bulky disease (defined as a tumor diameter >=10 cm), bone marrow infiltration and the IPI. (ii) Treatment: monotherapy with alkylating agents versus combination chemotherapies without adriamycin versus adriamycin-containing regimens. (iii) Response to therapy: CR versus PR versus no response. (iv) RD: defined as the time from response to relapse/progression (<2 versus >=2 years). (v) Variables at relapse: (a) clinical data (age, performance status, presence of B-symptoms); (b) histological subtype (FL versus transformed into high-grade lymphoma); (c) tumor extension data (bulky disease, Ann Arbor stage, extranodal involvement, bone marrow involvement, serum LDH and ß2M); (d) IPI; (e) salvage treatment; and (f) response to salvage treatment.

Statistical analysis
Overall survival was defined as the time between the date of diagnosis and the date of last follow-up or death; cause-specific survival, as the date of diagnosis to the date of last follow-up or lymphoma-related death; time to treatment failure (TTF), as the start of treatment to one of the following events: death from any cause, disease progression or relapse, or date of last observation [22]. Survival from progression (SFP) for responding patients (CR or PR) was defined as the time from relapse or progression to the date of last follow-up or death [23]. Cause-specific SFP for responding patients (CR or PR) was defined as the time from relapse or progression to the date of last follow-up or lymphoma-related death.

The main end point of the study was SFP. In addition, TTF and overall survival were analyzed. Actuarial survival analysis was performed according to the method described by Kaplan and Meier [24], and the curves compared by the log-rank test [25]. The univariate analysis was carried out for each of the parameters indicated above. All significant prognostic factors in the univariate study were included in a multivariate analysis performed by Cox’s stepwise proportional hazard regression method [26].


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Time to treatment failure and overall survival in the whole series
TTF in the overall series (201 patients) is plotted in Figure 1. Five-year TTF was 42% [95% confidence interval (CI) 34% to 49%].



View larger version (17K):
[in this window]
[in a new window]
 
Figure 1. Time to treatment failure (TTF) and overall survival (OS) in 201 patients with follicular lymphoma. Five-year TTF: 42% (95% CI 34% to 49%). Five-year OS: 71% (95% CI 65% to 76%).

 
Seventy-two patients died during the follow-up period. The median overall survival was 9.3 years, and the 5-year and 10-year overall survival were 71% (95% CI 65% to 76%) and 48% (95% CI 39% to 57%), respectively (Figure 1). Eleven patients (15%) died from non-lymphoma-related causes, with a median cause-specific survival of 14.3 years.

Characteristics of the patients at progression, salvage therapy and response
Of 164 responders, 90 (55%) presented with disease progression or relapse during the follow-up period. These patients had been registered from January 1976 to August 1997, with a median follow-up for alive patients of 6.25 years (range 2.1–21.6 years). The median age at progression was 56 years (range 26–78 years). The most important characteristics of these patients both at diagnosis and at progression are listed in Table 1. Among 82 assessable patients, 55 (67%) were in advanced stage and 35 (42%) had an intermediate- or high-risk prognosis, according to the IPI. Among 45 rebiopsed patients, histological transformation into a high-grade lymphoma was observed in nine cases (20%) (diffuse large-cell lymphoma in all of these). Patients in whom a new biopsy was not performed were older than those rebiopsed (patients >=65 years, 33% versus 13%, respectively; P = 0.04). Treatment at progression varied during the study period and consisted of: monotherapy with alkylating agents, 27 patients (30%); combination chemotherapy, 40 patients (45%); fludarabine alone or in combination, 12 patients (13%); and other treatments, 11 patients (12%) including only radiotherapy in three stage I patients. In addition, stem cell transplantation was performed in seven cases as part of the salvage treatment. In the subset of nine patients, in whom histological transformation was diagnosed, combination chemotherapy was given in all the cases, followed by autologous stem cell transplantation in four cases. Thirty-four (41%) of 83 assessable patients had a CR, 29 (35%) a PR, while 20 patients (24%) were considered as non-responders.

Survival from progression
The median SFP was 4.8 years, with a 5-year SFP of 47% (95% CI 35% to 58%) (Figure 2). The most important variables predicting SFP are detailed in Table 2. As indicated, advanced stage at diagnosis was the only initial factor retaining predictive value for SFP. First-line therapy did not influence SFP. In the univariate analysis, unfavorable prognostic factors at progression were bulky disease, poor performance status, high serum LDH level, high serum ß2M, bone marrow involvement, advanced stage, intermediate- or high-risk IPI, and an RD <2 years. Older age (>=60 years) showed a trend for statistical significance. Regarding salvage therapy, there were no differences in SFP according to the salvage regimen used, including ASCT. There was a trend for a higher SFP in patients receiving fludarabine combinations (P = 0.09), but these patients were highly selected in terms of age, performance status and follow-up. A relatively high number of missing data was found for four of these variables (LDH, ß2M, bone marrow involvement and transformation). We analyzed the SFP for patients with known data versus those with missing data. Patients with unknown LDH, bone marrow or lymph node biopsy had a worse SFP than the others, whereas no significant differences were found for ß2M (data not shown). The SFP according to the RD is shown in Figure 3. When the same analysis was performed in the subset of patients in advanced stage at progression, no significant differences were found according to the RD [5-year SFP: 21% (95% CI 0% to 44%) and 45% (95% CI 19% to 70%) for RD <2 and RD >=2 years, respectively; P = not significant]. In the subgroup of patients in localized stage, the RD maintained its prognostic significance with a 5-year SFP of 38% (95% CI 7% to 69%) in patients with a RD <2 years and a 5-year SFP of 92% (95% CI 78% to 100%) in patients with a longer RD (P = 0.013). When RD and stage were compared by a Cox regression, only stage retained prognostic significance for SFP [P = 0.015, relative risk (RR) 2.9 (95% CI 1.2–6.8)]. The univariate analysis was carried out for the subgroup of patients <60 years old, with this showing the same prognostic factors for SFP (data not shown). On the other hand, among the 45 patients in whom a new lymph node biopsy was obtained, no differences in survival were found according to the histological transformation event [5-year SFP 63% (95% CI 44% to 82%) and 62% (95% CI 29% to 95%) for transformed and non-transformed patients, respectively]. Non-biopsed patients had a shorter SFP than biopsed patients [5-year SFP 32% (95% CI 16% to 48%) versus 62% (95% CI 46% to 78%); P = 0.004].



View larger version (17K):
[in this window]
[in a new window]
 
Figure 2. Survival from progression (SFP) and cause-specific survival from progression (cause-specific SFP) in 90 responding patients who eventually progressed. Five-year SFP: 47% (95% CI 35% to 58%). Five-year cause-specific SFP 48% (95% CI 36% to 60%).

 

View this table:
[in this window]
[in a new window]
 
Table 2. Prognostic factors for survival from progression in 90 patients with follicular lymphoma who presented with relapse: results of the univariate analysis
 


View larger version (16K):
[in this window]
[in a new window]
 
Figure 3. Survival from progression according to the duration of response (RD) to first-line therapy. Five-year SFP 63% (95% CI 45% to 80%) and 33% (95% CI 17% to 48%), for patients with RD >=2 years and RD <2 years, respectively.

 
For the multivariate analysis all the variables with prognostic significance in the univariate study (P value <0.1) and with a sufficient number of assessable cases were selected. As a result, LDH level, ß2M level, bone marrow involvement and the histology at relapse were not included. On the other hand, response to initial treatment (CR versus PR) was included, due to its clinical interest. Finally, a multivariate analysis was performed including stage at diagnosis (I–II versus III–IV), initial clinical response (CR versus PR), age at relapse (continuous variable), stage at relapse (I–II versus III–IV), bulky disease at relapse, performance status at relapse (ECOG <2 versus ECOG >=2) and RD (<2 versus >=2 years). Seventy-nine of the 90 study patients were included in the final multivariate analysis for SFP. The factors retaining prognostic importance for SFP were: RD [P = 0.011, RR 0.37 (95% CI 0.17–0.8)] and poor performance status at progression [P <0.001, RR 9.0 (95% CI 3.6–22)]. As can be observed in Figure 4, only 10 of the 33 patients with ECOG <2 and RD >=2 died during the follow-up, with a 5-year SFP of 68% (95% CI 50% to 86%). Five-year SFP was 36% (95% CI 16% to 56%) for those patients having one of the two unfavorable variables, and 0% for those presenting the two unfavorable variables. Finally, in order to assess the prognostic value of the IPI in the setting of relapsed patients, we performed the multivariate analysis including the IPI at relapse, along with the above-mentioned variables. Poor performance status at progression [P = 0.0008; RR 5.09 (95% CI 1.9–13.3)], intermediate- or high-risk IPI at progression [P = 0.025; RR 2.5 (95% CI 1.1–5.5)] and an RD <2 years [P = 0.02; RR 2.4 (95% CI 1.14–5.3)] retained prognostic value for SFP when we included IPI at relapse in the multivariate analysis. In the subgroup of patients <60 years old, the multivariate analysis showed that performance status (P <0.001) and RD (P = 0.01), in addition to bulky disease at progression (P = 0.001), were the most important features for predicting SFP.



View larger version (19K):
[in this window]
[in a new window]
 
Figure 4. Survival from progression according to performance status (ECOG) and the response duration (RD) (P <0.00001). Five-year SFP 68% (95% CI 50% to 86%), 36% (95% CI 16% to 56%) and 0%, for patients with ECOG <2 and RD >=2, ECOG >=2 or RD <2, and patients with ECOG >=2 and RD <2, respectively.

 
Cause-specific survival from progression
Thirty-nine of 90 relapsed patients died from lymphoma, whereas three patients died from non-lymphoma related causes. Five-year cause-specific SFP was 48% (95% CI 36% to 60%) (Figure 2). A multivariate analysis was carried out following the above criteria. Performance status at relapse [P = 0.002; RR 5.1 (95% CI 1.8–14.1)] and RD [P = 0.04; RR 0.4 (95% CI 0.2–0.9)] emerged as the most important variables to predict cause-specific SFP.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
FL is among the most frequent forms of lymphomas in Western countries, where it accounts for 22% of all lymphomas. The median age of the patients at diagnosis was 59 years and the prognosis extremely variable, the median survival in recent series being up to 10 years [14, 27]. FL is incurable with standard therapies.

The prognostic factors for patients with FL at diagnosis have been extensively investigated. Thus, older age, poor performance status, advanced stage, bulky disease, leukemic expression, high serum LDH and ß2M levels, and a high risk according to the IPI have consistently been found to be associated with a poor outcome [1, 2, 2835]. The prognostic significance of many of these variables (i.e. age, stage and LDH) has been confirmed in our series (data not shown).

The natural history of FL is characterized by continuous relapses that become progressively less sensitive to chemotherapy [1]. Although patients with relapsed FL often still respond to further therapy, the RD becomes shorter after each relapse [1] and most patients die due to the disease. All of the above make the management of these patients challenging, with a wide variety of treatment options ranging from a watchful waiting policy to intensive therapies [1016, 3642].

New therapeutic approaches are being developed and investigated. These include purine analogs alone or in combination [5, 6], monoclonal antibodies [79], high-dose therapy followed by autologous or allogeneic stem-cell transplantation [1016], as well as different biological response modifiers. Preliminary data in selected patients receiving intensive regimens indicate that CR rates are high (up to 90%) and that such responses may be maintained for long periods [5, 11, 16]. However, no differences in terms of survival between conventional and more intensive treatments have been found [10]. Whereas experimental therapies are rarely used at diagnosis, with the exception of investigational trials, such approaches, particularly autologous stem cell transplantation, are increasingly being offered to patients in disease progression [1016]. However, patients in whom the disease progresses may still have a relatively long survival with conventional management [1]. On the other hand, the morbidity with some of the new therapies can be high and their long-term toxicity is still largely unknown. This is particularly true with autologous stem-cell transplants, which have been associated in recent analysis with an incidence of MDS/AML ranging from 4 to 12% [14, 15, 18]. Given all these facts, the analysis of prognostic factors at the time of the first relapse or progression could be useful in identifying those patients in whom, due to their extremely poor prognosis, experimental therapies are warranted.

Whereas the prognostic variables of patients with FL at diagnosis have been largely investigated [1, 2, 2835], the prognosis of patients in relapse or progression has been less studied. Advanced age, presence of B-symptoms, anemia and extensive extranodal involvement have been identified as unfavorable clinical characteristics for survival after progression [23, 43, 44]. In the present study, some clinical characteristics at relapse (stage, bulky, performance status and IPI) showed their prognostic value for SFP. Age, although showing a trend for statistical significance, did not reach prognostic value in the multivariate analysis. In fact, the median age in the present series is slightly lower than in others [1, 27, 42]. The number of previous therapies has also been found to be an important factor to predict survival [43, 44]. Finally, other studies [23, 45] showed the importance of the response to the initial therapy and its duration, as well as the response to salvage therapy in the survival of the patients. The prognostic impact of RD, a variable of well known importance in large-cell lymphoma as well as in other malignancies [46, 47], has only been investigated in a few studies. Weisdorf et al. [23] found age at relapse <60 years, CR and a RD >1 year to be associated with a longer survival after progression (5-year SFP was 43% for patients with a RD >=1 year, whereas it was 28% for patients with a shorter RD; P = 0.0002). In this study, however, no features at relapse, apart from age, were analyzed. Regarding salvage therapy, although there was a trend for a higher SFP among patients who received fludarabine combinations, these patients were younger and their follow-up shorter than the others. On the other hand, molecular response has also been demonstrated to correlate with outcome in FL patients [48], but this is rarely applicable in routine clinical practice. Moreover, the prognostic importance of molecular follow-up in FL patients is not yet well established.

One limitation of our study, as well as of other studies dealing with patients at relapse, is the difficulty in performing an adequate staging at the moment of relapse or progression. For instance, in our series a bone marrow biopsy was repeated in only 42 of 90 patients. This could lead to relative downstaging of the patients and, subsequently, to decreased IPI scores. Similarly, a new lymph node biopsy to rule out transformation into a high-grade lymphoma was performed in just half of the patients. The main reasons for this were the older age and poor performance status of some patients (inviting us to spare such procedures) and patient’s refusal in some cases. Therefore, although we did not find differences among transformed and non-transformed patients, the lack of adverse impact on SFP of histological transformation is most likely due to the low number of cases, and no firm conclusions can be drawn in this regard. On the other hand, those patients in whom no lymph node or bone marrow biopsies or LDH determination were performed showed an unfavorable SFP; this stresses that these procedures were omitted in older patients with poor performance status.

In the present study, RD emerged as an important prognostic factor, with a 5-year SFP ranging from 33% (RD <2 years) to 63% (RD >=2 years); this cut-off was chosen because it was close to the median RD (2.3 years) (Figure 3). In the multivariate analysis, poor performance status at relapse along with RD retained prognostic significance for survival. The same variables, as well as bulky disease at relapse retained prognostic significance for survival in the subset of younger (<60 years) patients. This indicates that not only the RD, but also clinical characteristics at progression, are critical to determining the outcome for these patients. Interestingly, patients with no unfavorable variables (ambulatory performance status and RD >=2 years) had an excellent outcome after progression with a 5-year survival of 68%. These patients should probably not be considered for experimental treatments, particularly those that are potentially toxic. On the other hand, the median SFP of those patients with any risk factor is <5 years and therefore they may be candidates for experimental and intensive regimens. Of note, the degree of response to first-line therapy (CR versus PR) had only marginal significance in the multivariate analysis both in the whole series and in the group of patients <60 years of age. However, this is in contrast to another study, in which RD and features at relapse were not analyzed together [23].

In conclusion, clinical characteristics at progression, as well as the duration of first response to therapy, are factors that can be useful to select patients with FL as candidates for experimental therapies and to determine the most appropriate timing for such procedures.


    Acknowledgements
 
This work was supported in part by grants (98/0996 and 99/0189) from the Fondo de Investigación Sanitaria, Spanish Ministry of Health, Spain, Generalitat de Catalunya (1997XT-00023) and 00/P-EM.


    Footnotes
 
+ Correspondence to: Dr A. López-Guillermo, Department of Hematology, Hospital Clínic, Villarroel 170, 08036-Barcelona, Spain. Tel: +34-93-227-5575; Fax: +34-93-227-5428; E-mail: alopezg@clinic.ub.es Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Gallagher CJ, Gregory WM, Jones AE et al. Follicular lymphoma: prognostic factors for response and survival. J Clin Oncol 1986; 4: 1470–1480.[Abstract]

2. Romaguera JE, McLaughlin P, North L et al. Multivariate analysis of prognostic factors in stage IV follicular low-grade lymphoma: a risk model. J Clin Oncol 1991; 9: 762–769.[Abstract]

3. Lister TA. The management of follicular lymphoma. Ann Oncol 1991; 2 (Suppl 2): 131–135.[ISI][Medline]

4. López-Guillermo A, Montserrat E, Bosch F et al. Low-grade lymphoma: clinical and prognostic studies in a series of 143 patients from a single institution. Leuk Lymphoma 1994; 15: 159–165.[ISI][Medline]

5. Keating MJ, McLaughlin P, Cabanillas F. Low-grade non-Hodgkin’s lymphoma—development of a new effective combination regimen (fludarabine, mitoxantrone and dexamethasone; FND). Eur J Cancer Care (Engl) 1997; 6 (4 Suppl): 21–26.[Medline]

6. Zinzani PL, Bendandi M, Magagnoli M et al. Fludarabine-mitoxantrone combination-containing regimen in recurrent low-grade non-Hodgkin’s lymphoma. Ann Oncol 1997; 8: 379–383.[Abstract]

7. McLaughlin P, Grillo-López AJ, Link BK et al. Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol 1998; 16: 2825–2833.[Abstract]

8. Gordon LI, White CA, Witzig TE et al. ZevalinTM (IDEC-Y2B8) radioimmunotherapy of Rituximab refractory follicular non-Hodgkin’s lymphoma (NHL): interim results. Blood 1999; 94 (Suppl 1): 91a (Abstr 396).

9. Vose JM, Zelenetz AD, Rohatiner A et al. Iodine I 131 tositumomab for patients with follicular non-Hodgkin’s lymphoma (NHL): overall clinical trial experience by histology. Blood 1999; 94 (Suppl 1): 89a (Abstr 387).

10. Rohatiner AZ, Johnson PW, Price CG et al. Myeloablative therapy with autologous bone marrow transplantation as consolidation therapy for recurrent follicular lymphoma. J Clin Oncol 1994; 12: 1177–1184.[Abstract]

11. Bastion Y, Brice P, Haioun C et al. Intensive therapy with peripheral blood progenitor cell transplantation in 60 patients with poor-prognosis follicular lymphoma. Blood 1995; 86: 3257–3262.[Abstract/Free Full Text]

12. Bierman PJ, Vose JM, Anderson JR et al. High-dose therapy with autologous hematopoietic rescue for follicular low-grade non-Hodgkin’s lymphoma. J Clin Oncol 1997; 15: 445–450.[Abstract]

13. Apostolidis J, Gupta RK, Grenzelias D et al. High-dose therapy with autologous bone marrow support as consolidation of remission in follicular lymphoma: long-term clinical and molecular follow-up. J Clin Oncol 2000; 18: 527–536.[Abstract/Free Full Text]

14. Micallef IN, Lillington DM, Apostolidis J et al. Therapy-related myelodysplasia and secondary acute myelogenous leukemia after high-dose therapy with autologous hematopoietic progenitor-cell support for lymphoid malignancies. J Clin Oncol 2000; 18: 947–955.[Abstract/Free Full Text]

15. Freedman AS, Neuberg D, Mauch P et al. Long-term follow-up of autologous bone marrow transplantation in patients with relapsed follicular lymphoma. Blood 1999; 94: 3325–3333.[Abstract/Free Full Text]

16. Verdonck LF, Dekker AW, Lokhorst HM et al. Allogeneic versus autologous bone marrow transplantation for refractory and recurrent low-grade non-Hodgkin’s lymphoma. Blood 1997; 90: 4201–4205.[Abstract/Free Full Text]

17. Cheson BD, Vena DA, Barrett J, Freidlin B. Second malignancies as a consequence of nucleoside analog therapy for chronic lymphoid leukemias. J Clin Oncol 1999; 17: 2454–2460.[Abstract/Free Full Text]

18. Krishnan A, Bhatia S, Slovak ML et al. Predictors of therapy-related leukemia and myelodysplasia following autologous transplantation for lymphoma: an assessment of risk factors. Blood 2000; 95: 1588–1593.[Abstract/Free Full Text]

19. Harris NL, Jaffe ES, Stein H et al. A revised European-American classification of lymphoid neoplasms: a proposal from the International Lymphoma Study Group. Blood 1994; 84: 1361–1392.[Free Full Text]

20. Harris NL, Jaffe ES, Diebold J et al. The World Health Organization classification of neoplasms of the hematopoietic and lymphoid tissues: report of the clinical advisory committee meeting-Airlie House, Virginia, November, 1997. Ann Oncol 1999; 10: 1419–1432.[Abstract]

21. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project. A predictive model for aggressive non-Hodgkin’s lymphoma: The International NHL Prognostic Factors Project. N Engl J Med 1993; 329: 987–994.[Abstract/Free Full Text]

22. Cheson BD, Horning SJ, Coiffier B et al. Report of an international workshop to standardize response criteria for non-Hodgkin’s lymphomas. NCI Sponsored International Working Group. J Clin Oncol 1999; 17: 1244–1253.[Abstract/Free Full Text]

23. Weisdorf DJ, Andersen JW, Glick JH, Oken MM. Survival after relapse of low-grade non-Hodgkin’s lymphoma: implications for marrow transplantation. J Clin Oncol 1992; 10: 942–947.[Abstract]

24. Kaplan GL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

25. Peto R, Pike MC. Conservatism of the approximation {Sigma}(O-E)2/E in the log-rank test for survival data or tumor incidence data. Biometrics 1973; 29: 579–584.[ISI][Medline]

26. Cox DR. Regression models and life tables. J R Stat Soc (B) 1972; 34: 187–220.[ISI]

27. The Non-Hodgkin’s Lymphoma Classification Project. A clinical evaluation of the International Lymphoma Study Group classification of non-Hodgkin’s lymphoma. Blood 1997; 89: 3909–3918.[Abstract/Free Full Text]

28. Cabanillas F, Smith T, Bodey GP et al. Nodular malignant lymphomas: factors affecting complete response rate and survival. Cancer 1979; 44: 1983–1989.[ISI][Medline]

29. Soubeyran P, Eghbali H, Bonichon F et al. Low-grade follicular lymphomas: analysis of prognosis in a series of 281 patients. Eur J Cancer 1991; 27: 1606–1613.[ISI][Medline]

30. Litam P, Swan F, Cabanillas F et al. Prognostic value of serum ß-2 microglobulin in low-grade lymphoma. Ann Intern Med 1991; 114: 855–860.[ISI][Medline]

31. Leonard RC, Hayward RL, Prescott RJ, Wang J-X. The identification of discrete prognostic groups in low grade non-Hodgkin’s lymphoma. The Scotland and Newcastle Lymphoma Group Therapy Working Party. Ann Oncol 1991; 2: 655–662.[Abstract]

32. Cameron DA, Leonard RC, Mao JH, Prescott RJ. Identification of prognostic groups in follicular lymphoma. The Scotland and Newcastle Lymphoma Group Therapy Working Party. Leuk Lymphoma 1993; 10: 89–99.[ISI][Medline]

33. Federico M, Vitolo U, Zinzani PL et al. Prognosis of follicular lymphoma: a predictive model based on a retrospective analysis of 987 cases. Intergruppo Italiano Linfomi. Blood 2000; 95: 783–789.[Abstract/Free Full Text]

34. Coiffier B, Bastion Y, Berger F et al. Prognostic factors in follicular lymphomas. Semin Oncol 1993; 20 (Suppl 5): 89–95.[ISI][Medline]

35. López-Guillermo A, Montserrat E, Bosch F et al. Applicability of the International Index for aggressive lymphomas to patients with low-grade lymphoma. J Clin Oncol 1994; 12: 1343–1348.[Abstract]

36. Horning SJ, Rosenberg SA. The natural history of initially untreated low-grade non-Hodgkin’s lymphomas. N Engl J Med 1984; 311: 1471–1475.[Abstract]

37. Case DC Jr. Comparison of M-2 protocol with COP in patients with nodular lymphoma. Oncology 1984; 41: 159–163.[ISI][Medline]

38. McLaughlin P, Fuller LM, Velasquez WS et al. Stage III follicular lymphoma: durable remissions with a combined chemotherapy-radiotherapy regimen. J Clin Oncol 1987; 5: 867–874.[Abstract]

39. Young RC, Longo DL, Glatstein E et al. The treatment of indolent lymphomas: watchful waiting versus aggressive combined modality treatment. Semin Hematol 1988; 25: 11–16.[ISI][Medline]

40. McLaughlin P, Fuller L, Redman J et al. Stage I-II low-grade lymphomas: a prospective trial of combination chemotherapy and radiotherapy. Ann Oncol 1991; 2 (Suppl 2): 137–140.[Abstract]

41. Mac Manus MP, Hoppe RT. Is radiotherapy curative for stage I and II low-grade follicular lymphoma? Results of a long-term follow-up study of patients treated at Stanford University. J Clin Oncol 1996; 14: 1282–1290.[Abstract]

42. Wood LA, Coupland RW, North SA, Palmer MC. Outcome of advanced stage low grade follicular lymphomas in a population-based retrospective cohort. Cancer 1999; 85: 1361–1368.[ISI][Medline]

43. Johnson PW, Rohatiner AZ, Whelan JS et al. Patterns of survival in patients with recurrent follicular lymphoma: a 20-year study from a single center. J Clin Oncol 1995; 13: 140–147.[Abstract]

44. Spinolo JA, Cabanillas F, Dixon DO et al. Therapy of relapsed or refractory low-grade follicular lymphomas: factors associated with complete remission, survival and time to treatment failure. Ann Oncol 1992; 3: 227–232.[Abstract]

45. Davidge-Pitts M, Dansey R, Bezwoda WR. Prolonged survival in follicular non Hodgkin’s lymphoma is predicted by achievement of complete remission with initial treatment: results of a long term study with multivariate analysis of prognostic factors. Leuk Lymphoma 1996; 24: 131–140.[ISI][Medline]

46. Guglielmi C, Gomez F, Philip T et al. Time to relapse has prognostic value in patients with aggressive lymphoma enrolled onto the Parma trial. J Clin Oncol 1998; 16: 3264–3269.[Abstract]

47. Kern W, Schoch C, Haferlach T et al. Multivariate analysis of prognostic factors in patients with refractory and relapsed acute myeloid leukemia undergoing sequential high-dose cytosine arabinoside and mitoxantrone (S-HAM) salvage therapy: relevance of cytogenetics abnormalities. Leukemia 2000; 14: 226–231.[ISI][Medline]

48. López-Guillermo A, Cabanillas F, McLaughlin P et al. The clinical significance of molecular response in indolent follicular lymphomas. Blood 1998; 91: 2955–2960.[Abstract/Free Full Text]