Can patients with early-stage diffuse large B-cell lymphoma be treated without bone marrow biopsy?

S. T. Lim*, M. Tao, Y. B. Cheung, S. Rajan and B. Mann

Department of Medical Oncology, National Cancer Centre, Singapore

* Correspondence to: Dr S. T. Lim, National Cancer Centre, 11 Hospital Drive, 169610 Singapore. Tel: +65-6436-8172; Fax: +65-6227-2759; Email: dmolst{at}nccs.com.sg


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background: Data on the incidence of bone marrow (BM) involvement in early-stage diffuse large B-cell lymphoma (DLBCL) are lacking. Although BM biopsy is a safe procedure, it is often poorly tolerated. This analysis aims to assess the incidence of BM involvement and to identify parameters predicting BM involvement in early-stage DLBCL.

Patients and methods: One hundred and ninety-two patients with radiological stages 1 and 2 disease were analysed. The data collected were age, sex, presence of B symptoms, white blood cell (WBC) count, platelet count, haemoglobin (Hb), serum lactate dehydrogenase level, serum ß2-microglobulin level, presence of extranodal disease, and the presence of bulky disease (defined as >7 cm).

Results: Overall incidence of BM involvement was 3.6%. Hb < 10 g/dl (P=0.02), WBC count <4 x 109/l (P=0.007) and bulky disease (P=0.06) were found to be predictive of BM involvement. Among the 120 patients without any of these three factors, only one patient had BM involvement (0.83%; 95% confidence interval 0.02% to 4.6%). The absence of all three factors gave a negative predictive value of 99.2%. Overall 3-year survival for patients without all three risk factors was 80%.

Conclusions: BM biopsy may be safely omitted in selected patients with early-stage DLBCL.

Key words: bone marrow biopsy, early-stage DLBCL, incidence, prognostic


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
In the evaluation of patients with newly diagnosed lymphoma, it is generally recommended that a bone marrow aspirate and trephine biopsy be done as part of the staging investigations. Nevertheless, the chance of finding bone marrow involvement varies considerably among the different subtypes of lymphoma: it is found in up to 50% of patients with follicular lymphoma and 70% of those with small lymphocytic lymphoma, but in only 15% of patients with diffuse large B-cell lymphoma (DLBCL) [1Go]. For patients with DLBCL, there is little available data regarding the incidence of bone marrow involvement in patients with limited evidence of disease. Of interest, one study of 244 such patients found that no patient with stage IA or stage IIA disease had bone marrow involvement [2Go].

Although bone marrow biopsy is considered a rather safe procedure, with low rates of complication, it is often a source of concern and fear to the patient [3Go, 4Go]. Thus, in a prospective study of 132 adults, 36% reported moderate to severe pain after the procedure [4Go]. It would be extremely useful to determine whether bone marrow biopsy and aspiration could safely be avoided in any particular group of patients with DLBCL.

Recently, specialised techniques such as flow cytometry, cytogenetics analysis, immunohistochemistry and detection of immunoglobulin gene rearrangement have been investigated to detect minimal histologic involvement not diagnosed by conventional methods. These tests, however, are expensive and the routine use of these tests in bone marrow staging in patients with early-stage DLBCL is uncertain. It may be more cost-effective to perform them in patients with clinical characteristics predicting a higher likelihood of marrow involvement.

The main purpose of this retrospective analysis was to describe the true incidence of bone marrow involvement in patients with early-stage DLBCL according to radiological and clinical criteria, and to identify clinical and laboratory parameters that might predict bone marrow involvement in patients with early-stage disease. We also hoped to define a subgroup of patients with early-stage DLBCL in whom bone marrow examination can be safely omitted during the initial staging work-up.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Three hundred and eighteen patients with DLBCL diagnosed between June 1996 and December 2002 were seen in the Department of Medical Oncology at the National Cancer Centre of Singapore. All patients underwent history, physical examination, standard laboratory tests, and computed tomography (CT) scans of chest, abdomen and pelvis. In addition, all patients underwent unilateral bone marrow aspirate and biopsy. Flow cytometry of marrow or cytogenetic analysis were only performed in some patients at the discretion of the treating physician. The pathology and clinical staging of all patients were centrally reviewed.

We confined the analysis to 192 patients who were found to have Ann Arbor stages I and II disease after clinical and CT scan staging alone. The following prospectively collected data were analysed: age, sex, presence of B symptoms, white blood cell (WBC) count, platelet (PLT) count, haemoglobin (Hb), serum lactate dehydrogenase level (LDH), serum ß2-microglobulin (B2M) level, presence of extranodal disease, presence of bulky disease (defined as tumour diameter >7 cm) and radiological stage. A low WBC count is defined as less than 4x109/l. A low Hb level was defined as less than 10 g/dl, while a low PLT count was defined as less than 100 x 109/l. An elevated LDH was defined as a value exceeding 380 U/l (our institution's upper normal limit), while an elevated B2M level was defined as greater than 1800 mmol/l (our institution's upper normal limit). Bone marrow was considered involved if either the aspirate or trephine biopsy showed the presence of lymphomatous infiltrate by standard morphologic assessment.

Patients with limited-stage DLBCL without bulky disease were treated with three to four cycles of cyclophosphamide, vincristine, adriamycin and prednisolone (CHOP) with involved field radiation (IFRT), while patients with bulky disease received six cycles of CHOP with IFRT. All patients with bone marrow involvement were treated with six to eight cycles of CHOP.

Statistical analysis
The association between bone marrow involvement and the following potential prognostic factors were assessed by the Fisher's exact test: age, bulky disease, low Hb level, low WBC count, low PLT count, B symptoms, elevated LDH, extranodal involvement and elevated B2M. Age, LDH and extranodal disease were chosen as they are considered important prognostic factors based on the International Prognostic Model [5Go]. Haematological parameters such as Hb level, WBC count and PLT count were chosen as they have been shown to be predictive of bone marrow involvement [6Go, 7Go]. Bulky disease, B symptoms and B2M level were chosen based on the premise that they are markers of tumour burden and were shown in other studies to be useful prognostic factors [8Go–10Go]. Statistical analyses were done by the STATA 7.0 software (Stata Corp, College Station, TX, USA). Fisher's exact test was used to evaluate the association of the various prognostic factors with bone marrow involvement. Confidence intervals (CIs) of prevalence were estimated using binomial distribution.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The clinical characteristics of the 192 patients with radiological stages 1 and 2 disease are summarized in Table 1.


View this table:
[in this window]
[in a new window]
 
Table 1. Characteristics of patients (n=192)

 
Overall, bone marrow involvement was found in seven patients (3.6%). Comparing the clinical characteristics between patients with and without bone marrow involvement, we showed that low Hb (P=0.02), low WBC (P=0.007) and bulky disease (P=0.06) are associated with bone marrow involvement (Table 2). Among the 120 patients without bulky disease, a Hb level of more than 10 g/dl and a WBC count of more than 4 x 10 9/l (Table 3), only one patient had bone marrow involvement (0.83%; 95% CI 0.02% to 4.6%).


View this table:
[in this window]
[in a new window]
 
Table 2. Factors predicting bone marrow involvement

 

View this table:
[in this window]
[in a new window]
 
Table 3. Incidence of bone marrow involvement according to risk factors

 
Six of the seven patients with bone marrow involvement had at least one of the three risk factors (low Hb, a low WBC count or bulky disease). The prevalence (95% CI) of bone marrow involvement in patients without any of the three major factors was 0.0083 (0.0002–0.0456), while that in patients with at least one of the three factors was 0.0833 (0.0312–0.1726). Hence, the presence of any one of these three risk factors in a patient gave a positive predictive value of 8.3% for bone marrow involvement, while the absence of all three factors gave a negative predictive value of 99.2% for bone marrow involvement (see Table 3).

The median follow-up for this group of patients was 2.6 years. The overall 3-year survival was 74%. For the group of patients with stage I or II non-bulky disease and normal Hb level and WBC count, the overall 3-year survival was 80%.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
We have demonstrated that the overall incidence of bone marrow involvement was quite low in patients with radiological stage I or II disease (3.6%). Furthermore, in patients with non-bulky disease (a Hb level >10 g/dl and a WBC count >4 x 109/l) the likelihood of marrow involvement was even lower, at 0.8%. The absence of these three factors gave a negative predictive value for bone marrow involvement by conventional methods in excess of 99%, suggesting that bone marrow examination could safely be avoided in this subgroup of patients. It should be noted, however, that bulky disease was only a borderline significant risk factor.

The observed overall 3-year survival rate of 73.4% (including patients with bulky disease) and 80% for patients with non-bulky disease with normal haematological parameters were very similar to the survival rates observed in patients with low-risk disease according to the International Prognostic Model [5Go]. Thus, it is unlikely that we had significantly ‘under-staged’ our patients, which would have resulted in a much poorer clinical outcome.

The potential advantage of bilateral versus unilateral bone marrow biopsy has been controversial. In one series of 176 patients with non-Hodgkin's lymphoma (NHL), the second bone marrow biopsy resulted in an upstaging in 2.5% of the patients with aggressive NHL [11Go]. Thus, bilateral bone marrow biopsy would appear to have a limited effect on upstaging most patients from stage I/II to stage IV [12Go].

Data on the routine use of flow cytometry to increase the sensitivity and specificity of bone marrow biopsy, particularly for DLBCL, are also scant [13Go, 14Go]. In a study by Palacio et al., flow cytometry was found to be similar to routine histological examination in the detection of bone marrow involvement in 65 patients with follicular large or DLBCL [13Go]. In a recent study of 155 patients with DLBCL, a significant difference in survival was noted between patients with negative bone marrow histology and bone marrow negative PCR results for immunoglobulin gene rearrangement versus those with negative bone marrow histology but positive PCR results [14Go]. The likelihood of detecting these molecular changes was significantly lower in patients with early-stage disease (10 of 85 patients or 11.7%) compared with patients with advanced stage (25 of 70 patients or 35.7%). Among patients with an International Prognostic Index (IPI) score of 0 to 1, the likelihood of detecting immunoglobulin gene rearrangement was 8% (six of 71 patients). The chance of a positive PCR analysis of marrow may be even lower in patients with limited disease without any poor prognostic features.

As approximately 50% of patients with DLBCL present as early-stage disease [15Go], it may not be cost-effective to perform these specialised molecular tests for all patients with normal bone marrow exams on histologic grounds. Instead, it may be more cost-effective to perform these tests for patients with limited disease and morphologically normal marrow but have clinical characteristics that predict for a greater likelihood of marrow involvement, such as the presence of a low WBC count as shown in this study.

The Southwest Oncology Group (SWOG) trial randomised a group of 401 patients with limited-stage aggressive lymphoma to receive either eight cycles of CHOP or three cycles followed by IFRT [16Go]. In contrast to the initial report that showed an advantage for abbreviated CHOP with IFRT, long-term follow-up of this trial has recently shown an increase in late relapses and lymphoma-related deaths in patients receiving short-course chemotherapy and IFRT [17Go]. Nevertheless, the risk of relapse appears to be related to the patient's score on the modified IPI scale. In patients with one or more poor prognostic factors on the modified IPI scale, the 5- and 10-year survivals declined to less than 70%. Even so, patients with no poor risk factors on the modified IPI scale treated with short-course CHOP and IFRT continued to do well. The 5- and 10-year survival in this group of patients were 95% and 90%, respectively, suggesting that in this selected subgroup the risk of occult systemic involvement was very low. This is consistent with the low incidence of bone marrow involvement in the select group of patients in our study.

A patient with clinically limited, non-bulky disease without a low Hb level or low WBC count has a very low risk of bone marrow involvement and may be treated without bone marrow staging. If such a patient has no poor risk factors on the modified IPI scale, abbreviated CHOP with IFRT remains an option. As shown in the SWOG study, for all other patients with one or more poor prognostic factors, regardless of marrow status, we would now recommend six to eight cycles of systemic chemotherapy.

With this recommendation, we believe that the decision to avoid bone marrow biopsy in selected patients will not compromise adequate therapy. We do recognise, however, that this is an explorative single institutional study involving only 192 patients with seven cases of bone marrow involvement, and the results should therefore be confirmed by other studies.

In conclusion, we have shown that the incidence of bone marrow involvement among patients with radiological early-stage DLBCL is 3.6%. Among patients without bulky disease with both a normal Hb and WBC count, the risk of bone marrow involvement is 0.8%. Among patients with any of these three risk factors, the risk of bone marrow involvement is 8.3%. This information is useful in helping clinicians decide the subgroup of patients where routine biopsy may be safely omitted, and identify the subgroup in which specialised tests to detect minimal morphological changes may be more cost-efficient.

Received for publication June 17, 2004. Revision received August 26, 2004. Accepted for publication September 23, 2004.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. 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]

2. Ajarim D, El Foudeh M, Rahal M et al. The role of bone marrow biopsy in non-Hodgkin's lymphoma in adult patients. Proc Am Soc Clin Oncol 2003; 22: 2436 (Abstr).

3. Bain BJ. Bone marrow biopsy morbidity and mortality. BMJ 2003; 121: 949–951.

4. Vanhelleputte P, Nijs K, Delforge M et al. Pain during bone marrow aspiration: prevalence and prevention. J Pain Symptom Manage 2003; 26: 860–886.[CrossRef][ISI][Medline]

5. Shipp MA, Harrington DP, Anderson JR et al. A predictive model for aggressive NHL: the International Non-Hodgkin's Lymphoma Prognostic Factors Project. N Engl J Med 1992; 329: 987–994.[CrossRef][ISI]

6. Bloomfield CD, McKenna RW, Brunning RD. Significance of haematological parameters in the non-Hodgkin's malignant lymphomas. Br J Haematol 1976; 1: 41–46.

7. Conlan MG, Armitage JO, Bast M et al. Clinical significance of hematologic parameters in non-Hodgkin's lymphoma at diagnosis. Cancer 1991; 5: 1389–1395.

8. Swan F Jr, Velasquez WS, Tucker S et al. A new serologic staging system for large-cell lymphomas based on initial beta 2-microglobulin and lactate dehydrogenase levels. J Clin Oncol 1989; 7: 1518–1527.[Abstract]

9. Mackintosh JF, Cowan RA, Jones M et al. Prognostic factors in stage I and II high and intermediate grade non-Hodgkin's lymphoma. Eur J Cancer Clin Oncol 1988; 10: 1617–1622.[CrossRef]

10. Cowan RA, Jones M, Harris M et al. Prognostic factors in high and intermediate grade non-Hodgkin's lymphoma. Br J Cancer 1989; 2: 276–282.

11. Ebie N, Loew JM, Gregory SA. Bilateral trephine bone marrow biopsy for staging non-Hodgkin's lymphoma—a second look. Hematol Pathol 1989; 3: 29–33.[Medline]

12. Luoni M, Declich P, De Paoli A et al. Bone marrow biopsy for the staging of non-Hodgkin's lymphoma: bilateral or unilateral trephine biopsy? Tumori 1995; 81: 410–413.[ISI][Medline]

13. Palacio C, Acebedo G, Navarrete M et al. Flow cytometry in the bone marrow evaluation of follicular and diffuse large B-cell lymphomas. Haematologica 2001; 9: 934–940.

14. Mitterbauer-Hohendanner G, Mannhalter C, Winkler K et al. Prognostic significance of molecular staging by PCR-amplification of immunoglobulin gene rearrangements in diffuse large B-cell lymphoma (DLBCL). Leukemia 2004; 6: 1102–1107.[CrossRef]

15. Moller MB, Pedersen NT, Christensen BE. Diffuse large B-cell lymphoma: clinical implications of extranodal versus nodal presentation—a population-based study of 1575 cases. Br J Haematol 2004; 2: 151–159.[CrossRef]

16. Miller T, Dahlberg S, Cassady J et al. Chemotherapy alone compared with chemotherapy plus radiotherapy for localized intermediate- and high-grade non-Hodgkin's lymphoma. N Engl J Med 1998; 339: 21–26.[Abstract/Free Full Text]

17. Miller TP, Leblanc M, Spier C et al. CHOP alone compared to CHOP plus radiotherapy for early-stage aggressive non-Hodgkin's lymphomas: update of the Soutwest Oncology Group (SWOG) randomized trial. Blood 2001; 98: 724a (Abstr).