Prognostic indexes in follicular lymphoma: a comparison of different prognostic systems

G. Perea1,*, A. Altés1, S. Montoto2, A. López-Guillermo2, E. Domingo-Doménech3, A. Fernández-Sevilla3, J. M. Ribera4, J. Grau4, C. Pedro5, J. Angel Hernández6, C. Estany7, J. Briones1, R. Martino1, A. Sureda1, J. Sierra1 and E. Montserrat2

1 Clinical Hematology Division, Hospital de la Santa Creu i Sant Pau, Autonomous University of Barcelona, Barcelona; 2 Hematology Department, Institut Clínic de Malalties Hemato-oncológiques, Hospital Clínic, IDIBAPS, University of Barcelona, Barcelona; 3 Hematology Department, Hospital Durán i Reynals, Institut Català d'Oncologia, Ciutat Sanitaria i Universitaria de Bellvitge (Hospitalet de Llobregat); 4 Clinical Hematology Division, Hospital Germans Trias i Pujol, Badalona; 5 Clinical Hematology Division, Hospital del Mar, Barcelona; 6 Hematology Department, Hospital de Mataró, Mataró; 7 Hematology Department, Hospital Mútua de Terrassa, Terrassa, Spain

* Correspondence to: Dr G. Perea, Department of Hematology, Hospital de la Santa Creu i Sant Pau, C/ Sant Antoni Maria Claret 167, 08025 Barcelona, Spain. Tel: +34-93-291-93-96; Fax: +34-93-291-94-66; Email: gperea{at}hsp.santpau.es


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background: The International Prognostic Index (IPI), initially designed for aggressive lymphomas, is also used in follicular lymphoma (FL) and other indolent lymphomas. Two new prognostic indexes have recently been proposed for FL [the Italian Lymphoma Intergroup (ILI) Index and the Follicular Lymphoma International Prognostic Index (FLIPI)].

Patients and methods: Three indexes, IPI [age >60 years, extranodal involvement two or more sites, elevated lactate dehydrogenase (LDH), Eastern Cooperative Oncology Group performance status ≥2, stage ≥3], ILI (age >60 years, extranodal involvement two or more sites, elevated LDH, male sex, B symptoms, erythrocyte sedimentation rate ≥30 mm first hour) and FLIPI (age >60 years, stage ≥3, elevated LDH, nodal involvement five or more, haemoglobin level ≤12 g/dl) were calculated in 411 patients with FL.

Results: Overall concordance between the three indexes was 54%. A total of 126 (31%) patients were included in the high-risk group according to IPI, 131 (32%) according to ILI and 157 (38%) after FLIPI application. Ten-year overall survival rates after applying the prognostic indexes (IPI, ILI and FLIPI) were, respectively: 72%, 71% and 72%, in the low-risk group; 51%, 60% and 49% in the intermediate-risk group; and 24%, 16% and 31% in the high-risk group.

Conclusions: In this series, all three indexes, IPI, ILI and FLIPI, were useful to classify FL patients into differentiated risk groups, although the FLIPI identified a larger proportion of high-risk patients than the IPI and ILI.

Key words: follicular lymphoma, prognostic index, survival


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Follicular lymphoma (FL) is most commonly seen in middle-aged patients and accounts for ~30% of newly diagnosed non-Hodgkin lymphomas (NHL) [1Go, 2Go]. Despite the prolonged median survival time, ~10 years, progression-free survival (PFS) and overall survival (OS) are poor in some patients. Several attempts to build up a prognostic index that is useful to make risk-adapted treatment recommendations have been made [2Go–11Go]. The International Prognostic Index (IPI) has been successfully applied to patients with FL, but it seems to have a limited discriminating power as most patients are allocated to the favourable- or the intermediate-risk groups [12Go, 13Go]. In the last few years two specific prognostic scores have been proposed: the Italian Lymphoma Intergroup Index (ILI) and, more recently, by an international group, the Follicular Lymphoma International Prognostic Index (FLIPI) [14Go, 15Go]. The IPI and ILI indexes have been applied in FL patients with different success [16Go, 17Go]. The aim of this study was to apply all three prognostic indexes in a large group of patients with FL and to try to ascertain the relative merits of each of them.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients’ characteristics
Four hundred and sixty-five patients with a histologically confirmed diagnosis of FL grade I or II according to the WHO classification, consecutively diagnosed at five hospitals from Barcelona between January 1976 and December 2001, were initially considered for this study. Data needed to calculate all three indexes were finally obtained from 411 patients (data from 72 patients had been also used in the initial FLIPI study [15Go]). Patients’ clinical characteristics at diagnosis are shown in Table 1. The median follow-up of surviving patients was 73 months (range 6–292). Stage at diagnosis was determined in all patients by clinical evaluation, chest and abdomen computed tomography and bone marrow biopsy. Patients received varying first-line treatments: 18 patients (4%) did not receive any treatment, since a watch-and-wait policy was adopted; 36 patients (10%) were treated with radiotherapy and/or surgery alone; 58 patients (14%) with a single alkylating agent (cyclophosphamide or chlorambucil); 51 patients (12%) with a combination chemotherapy regimen without an anthracycline (basically cyclophosphamide, vincristine and prednisone); 21 patients (5%) were treated with fludarabine combinations; and 227 patients (55%) with a chemotherapy regimen with an antracycline (CHOP/CNOP). Response to treatment was assessed within 3 months after therapy was completed. In general patients were evaluated 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 thereafter. Follow-up manoeuvres consisted of physical examination, routine blood test and radiologic studies of areas with initial disease. Response after treatment was available in most patients: 190 (49%) achieved a complete response (CR) with initial therapy and 149 (39%) a partial response (PR), whereas 45 patients (12%) failed to respond to treatment.


View this table:
[in this window]
[in a new window]
 
Table 1. Clinical characteristics of patients

 
Prognostic indexes calculation
The IPI was calculated according to the International Non-Hodgkin's Lymphoma Prognostic Factors Project [11Go]. The variables used were age (≤60 versus >60 years), performance status (Eastern Cooperative Oncology Group performance status 0 or 1 versus ≥2), Ann Arbor stage (I–II versus III–IV), extranodal involvement (less than two versus two or more sites) and serum lactate dehydrogenase (LDH) level (normal versus high). Three risk groups were defined by IPI: score 0–1, low-risk; score 2, intermediate-risk; score ≥3, high-risk. The high-intermediate and high-risk groups were joined to form a single high-risk group for comparisons with the other indexes.

The ILI index was calculated as detailed by the Italian Lymphoma Intergroup [14Go]. Six variables were used to construct this index, three of them also being included in IPI (age, extranodal involvement and LDH level). The other three variables considered were presence of B-symptoms, male sex and erythrocyte sedimentation rate ≥30 mm/h. Depending on the number of adverse prognostic factors (0–1, 2 or ≥3), patients were classified into low-, intermediate- or high-risk groups.

The FLIPI was calculated according to the Follicular Lymphoma International Prognostic Project [15Go]. The variables used to classify patients according to the FLIPI index were age ≥60, advanced stage (III–IV), increased serum LDH, haemoglobin level <12 g/dl and nodal involvement (five or more sites). Three risk groups were considered: score 0–1, low-risk; score 2, intermediate-risk; and score ≥3, high-risk.

Statistical analysis
CR was defined as the disappearance of tumour masses and disease-related symptoms, as well as normalisation of the initially abnormal tests and/or biopsies, for at least 1 month. PR was considered when measurable lesions decreased by at least 50%. Disease was considered stable when there was no clinical response or evidence of progression. Disease progression during or after treatment was also taken into account [18Go].

OS and PFS curves were calculated for each risk category according to the Kaplan–Meier method. OS was calculated from the date of diagnosis to the last follow-up or death regardless of the cause. PFS was calculated for all treated patients from the onset of therapy until disease progression, relapse or death. Survival curves were compared using the log-rank test. All data were analysed using Statistical Package for the Social Sciencies software (SPSS®). The limit of statistical significance for all analyses was defined as P ≤ 0.05.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Prognostic indexes application
The distribution of the patients according to the IPI, ILI and FLIPI indexes is detailed in Table 2. Overall concordance between the three classification systems was 54%: 221 patients were allocated to the same risk group with all three indexes. Concordance was 37% for low-risk groups, 10% for intermediate-risk groups and 36% for high-risk groups. The number of patients included in the high-risk group was 123 (31%), 131 (32%) and 157 patients (38%) after application of the IPI, ILI and FLIPI, respectively.


View this table:
[in this window]
[in a new window]
 
Table 2. Prognostic indexes and patient distribution according to risk in 411 patients with follicular lymphoma

 
When considering all risk groups, concordance between the IPI and ILI was 73%, and was 71% between the IPI and the FLIPI. Concordance for high-risk groups was 69% and 67%, respectively.

Of patients aged 60 years or younger (n=249), 36 (14%) were classified in the high-risk group according to the IPI, 39 (16%) according to the ILI and 50 (20%) according to the FLIPI.

Survival
Survival data for the whole population and for each risk group are summarised in Table 3, and OS and PFS curves according to the IPI, ILI and FLIPI are shown in Figures 1, and 3, respectively. Using the IPI system, three groups of patients with statistically different OS and PFS were distinguished. The 5- and 10-year OS rates were 88% and 72%, respectively, in the low-risk group, 78% and 51% in the intermediate-risk group, and 47% and 24% in the high-risk group (log-rank test 71.7; P <0.0001). Five-year PFS was 53% for the low-risk group, 31% for the intermediate-risk group and 17% for the high-risk group (log-rank test 53.5; P <0.0001).


View this table:
[in this window]
[in a new window]
 
Table 3. Survival data for all patients and risk groups

 


View larger version (18K):
[in this window]
[in a new window]
 
Figure 1. Curves according to International Prognostic Index (IPI). Overall survival (OS) and progression-free survival (PFS) curves according to IPI risk groups. L-R, low-risk; I-R, intermediate-risk; H-R, high-risk.

 


View larger version (18K):
[in this window]
[in a new window]
 
Figure 3. Curves according to Follicular Lymphoma International Prognostic Index (FLIPI) index. Overall survival (OS) and progression-free survival (PFS) curves according to FLIPI risk groups. L-R, low-risk; I-R, intermediate-risk; H-R, high-risk.

 
The ILI index also defined three groups of patients. The OS at 5 and 10 years from diagnosis for each ILI risk group was 90% and 71%, respectively, in the low-risk group, 78% and 60% in the intermediate-risk group and 45% and 16% in the high-risk group (log-rank test 100; P <0.0001). The 5-year PFS for treated patients was 50% for patients at low-risk, 36% for patients at intermediate-risk and 16% for patients at high-risk (log-rank test 68.3; P <0.0001). Although differences in survival probabilities between low- and intermediate-risk groups were observed, these did not reach statistical significance (P=0.08 for OS; and P=0.1 for PFS).

Finally, according to the FLIPI, the 5- and 10-year OS rates were 89% and 72% for low-risk patients, 78% and 49% for the intermediate-risk group and 54% and 31% for the high-risk group, respectively (log-rank test 48.6; P <0.0001). The 5-year PFS for the different risk groups defined by the FLIPI were 50%, 37% and 22%, respectively (log-rank test 32.5; P <0.0001).

No differences were observed between treatments received by patients included in each high-risk group according to the three prognostic indexes. Fifty-five per cent (n=227) of FL patients in this series were treated uniformly with chemotherapy regimens including an anthracycline (CHOP/CNOP), and 28% of these patients were included in the high-risk group according to the IPI, 33% according to the ILI and 38% according to the FLIPI. The 5-year OS and PFS rates were 74% and 38%, respectively. No differences were observed in OS and PFS rates when these patients were classified according to the three prognostic indexes.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Several attempts have been made to design predictive models for patients with FL, the aim being to identify patients in whom aggressive, experimental therapies are warranted. In most studies, older age, poor performance status, advanced stage, B-symptoms, bulky disease, bone marrow involvement, increased serum LDH and high serum ß2-microglobulin levels have been parameters consistently associated with survival [2Go, 19Go–22Go]. Although prognostic models based on these factors have been proposed [8Go–11Go], none of them has gained wide acceptance.

In this report we have compared the three prognostic indexes for FL most commonly employed (i.e. the IPI, ILI and FLIPI) in an attempt to determine the merits of each one of these prognostic models. The IPI, initially designed for use in aggressive lymphomas, is easily applicable in clinical practice [11Go], and is also valuable in low-grade lymphomas [12Go, 13Go]. A major setback with the IPI is that only a small percentage (~8–11%) of patients with FL are included in the high-risk group. The ILI and FLIPI indexes, specifically designed for FL, also include variables that are easy to calculate, and separate patients into different risk groups. Both the IPI and the ILI have been applied in a series of FL patients. Maartense et al. [16Go] found that the ILI index fitted grade I–II FL patients better, while the IPI showed a better discrimination among grade III FL patients. In our experience, we did not observe significant differences regarding the discriminating value between IPI and ILI indexes in a series of grade I–II FL patients [17Go].

Regarding the FLIPI, patients from the current series were distributed in three different survival groups, with an OS probability at 5 and 10 years very similar to those previously showed by Solal-Céligny et al. [15Go], although in our series a higher proportion of patients were included in the high-risk group (38% versus 27%). The FLIPI seems to classify a larger number of patients into the high-risk group than IPI and ILI, even when only younger patients (≤60 years old) are considered: 14%, 16% and 20% of patients were included in the high-risk group after applying IPI, ILI and FLIPI indexes, respectively. According to these results, all three systems are useful to distribute FL patients into different risk groups, although the ILI index was especially valuable to separate high-risk patients, because in contrast with the other two indexes, differences in survival between low-risk and intermediate-risk patients were less significant. Of note, median survival of patients in the high-risk group is quite long (30–34 months) whatever the prognostic system employed. Because of this, it could be argued that prognostic indexes for FL, as currently devised, are not useful for making treatment decisions, particularly if the therapy proposed conveys important morbo-mortality. The prognostic assessment of patients with FL could be improved by using other variables [23Go–27Go]. In addition, the prognostic impact of genomic aberrations in patients with FL has also been investigated and a negative impact has been described for deletions of 6q [28Go, 29Go]. The whole-genome microarray analysis of gene expression has also been applied in FL and a survival predictor model constructed according to the gene expression signatures derived from non-malignant immune cells presents in the tumour at diagnosis [30Go]. These biological findings may contribute significantly to risk assessment in patients with FL.

In conclusion, all three prognostic systems investigated (IPI, ILI and FLIPI) were useful to identify patients with FL and different survival probabilities. In our series, however, the FLIPI identified a larger number of patients in the high-risk category. Future research should aim at improving the prognostic assessment of patients with FL by combining clinical variables with recently discovered biological variables.



View larger version (18K):
[in this window]
[in a new window]
 
Figure 2. Curves according to Italian Lymphoma Intergroup (ILI) index. Overall survival (OS) and progression-free survival (PFS) curves according to ILI risk groups. L-R, low-risk; I-R, intermediate-risk; H-R, high-risk.

 

    Acknowledgements
 
This work was supported in part by grants from Instituto de Salud Carlos III (Red de Grupos de Investigación en Linfomas, G07/179, and Red Temática de Investigación Cooperativa de Centros de Cáncer, C03/10) and Fundación la Caixa (JS 2002–2004).

Received for publication February 21, 2005. Revision received April 18, 2005. Accepted for publication April 20, 2005.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Armitage JO, Weisenburger DD, for the Non Hodgkin's Lymphoma Classification Project. New approach to classify non-Hodgkin's lymphomas: clinical features of the major histologic subtypes. J Clin Oncol 1998; 16: 2780–2795.[Abstract/Free Full Text]

2. Decaudin D, Lepage E, Brousse N et al. Low-grade stage III–IV follicular lymphoma: multivariate analysis of prognostic factors in 484 patients—A study of the Groupe d'Etude des Lymphomes de l'Adulte. J Clin Oncol 1999; 17: 2499–2505.[Abstract/Free Full Text]

3. 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/Free Full Text]

4. McLaughlin P, Hagemeister FB, Romaguera JE et al. Fludarabine, mitoxantrone and dexamethasone: an effective new regimen for indolent lymphomas. J Clin Oncol 1996; 14: 1262.[Abstract/Free Full Text]

5. López R, Martino R, Sureda A et al. Autologous stem cell transplantation in advanced follicular lymphoma. A single center experience. Haematologica 1999; 84: 350–355.[ISI][Medline]

6. Mahé B, Milpied N, Mellerin MP et al. PCR detection of residual Bcl-2/IgH-positive cells after high-dose therapy with autologous stem cell transplantation is a prognostic factor for event-free survival in patients with low-grade follicular non-Hodgkin's lymphoma. Bone Marrow Transplant 2003; 31: 467–473.[CrossRef][ISI][Medline]

7. Seymour JF, Pro B, Fuller LM et al. Long-term follow-up of a prospective study of combined modality therapy for stage I–II indolent non-Hodgkin's lymphoma. J Clin Oncol 2003; 21: 2115–2122.[Abstract/Free Full Text]

8. Leonard RCF, Hayward RL, Prescott RJ, Wang JX. The identification of discrete prognostic groups in low-grade non-Hodgkin's lymphoma. Ann Oncol 1991; 2: 655–667.[Abstract]

9. 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/Free Full Text]

10. Cameron DA, Leonard RCF, Jian-Hua M, Prescott RJ. Identification of prognostic groups in follicular lymphoma. Leuk Lymph 1993; 10: 89–99.[ISI][Medline]

11. The Non-Hodgkin's Lymphoma Prognostic Factors Project. Development of a predictive model for aggressive lymphoma: The International NHL Prognostic Factors Project. N Engl J Med; 329 1993; 987–994.

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

13. Foussard C, Desablens B, Sensebe L et al. Is the International Prognostic Index for aggressive lymphomas useful for low-grade lymphoma patients? Applicability to stage III–IV patients. The GOELAM Group, France. Ann Oncol 1997; 8 (Suppl 1): 49–52.[Abstract]

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

15. Solal-Céligny P, Roy P, Colombat P et al. Follicular lymphoma international prognostic index. Blood 2004; 104: 1258–1265.[Abstract/Free Full Text]

16. Maartense E, le Cessie S, Kluin-Nelemans HC et al. Age-related differences among patients with follicular lymphoma and the importance of prognostic scoring systems: analysis from a population-based non-Hodgkin's lymphoma registry. Ann Oncol 2002; 13: 1275–1284.[Abstract/Free Full Text]

17. Perea G, Altés A, Montoto S et al. International and Italian prognostic indices in follicular lymphoma. Haematologica 2003; 88: 649–653.

18. 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]

19. 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]

20. Bastion Y, Berger F, Bryon PA et al. Follicular lymphomas: assessment of prognostic factors in 127 patients followed for 10 years. Ann Oncol 1991; 2 (Suppl 2): 123–129.[Abstract]

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

22. 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.[CrossRef][ISI][Medline]

23. López-Guillermo A, Cabanillas F, McDonnell TI et al. Correlation of Bcl-2 rearrangement with clinical characteristics and outcome in indolent follicular lymphoma. Blood 1999; 93: 3081–3087.[Abstract/Free Full Text]

24. López-Guillermo A, Cabanillas F, McLaughlin P et al. Molecular response assessed by PCR is the most important factor predicting failure-free survival in indolent follicular lymphoma: update of the MDACC series. Ann Oncol 2000; 11 (Suppl 1): 137–140.[CrossRef][Medline]

25. Zhao WL, Daneshpouy ME, Mounier N et al. Prognostic significance of bcl-xL gene expression and apoptotic cell counts in follicular lymphoma. Blood 2004; 103: 695–697.[Abstract/Free Full Text]

26. Bilalovic N, Kirsti Blystod A, Golouh R et al. Expression of bcl-6 and CD10 protein is associated with longer overall survival and time to treatment failure in follicular lymphoma. Am J Clin Pathol 2004; 121: 34–42.[CrossRef][ISI][Medline]

27. Akasaka T, Lossos IS, Levy R. BCL-6 gene translocation in follicular lymphoma: a harbinger of eventual transformation to diffuse aggressive lymphoma. Blood 2003; 102: 1931–1932.[Free Full Text]

28. Viardot A, Möller P, Högel J et al. Clinicopathologic correlations of genomic gains and losses in follicular lymphoma. J Clin Oncol 2002; 20: 4523–4530.[Abstract/Free Full Text]

29. Tilly H, Rossi A, Stamatoullas A et al. Prognostic value of chromosomal aberrations in follicular lymphoma. Blood 1994; 84: 1043–1049.[Abstract/Free Full Text]

30. Dave S, Wright G, Tan B et al. Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells. N Engl J Med 2004; 351: 2159–5169.[Abstract/Free Full Text]





This Article
Abstract
Full Text (PDF)
All Versions of this Article:
16/9/1508    most recent
mdi269v1
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Perea, G.
Articles by Montserrat, E.
PubMed
PubMed Citation
Articles by Perea, G.
Articles by Montserrat, E.