Rethinking Bone Marrows: Are They Necessary in Early Stage Diffuse B-Cell Lymphomas?

Abstract & Commentary

Synopsis: In a retrospective analysis of approximately 200 patients with early stage B-cell diffuse lymphoma, bone marrow aspirate and biopsy revealed involvement in only 3.6%. A review of a number of risk factors indicated 3 (low hemoglobin, low white blood count, and bulky disease) as predictors of marrow involvement. In fact, the absence of these 3 factors gave a negative predictive value of 99.2% for marrow involvement. The study concludes that bone marrow sampling may be safely omitted in early stage diffuse B-cell lymphoma in selected patients (those without the 3 mentioned risk factors).

Source: Lim ST, et al. Ann Oncol. 2005;16:215-218.

In the evaluation of newly diagnosed lymphoma, bone marrow aspirate and biopsy are typically performed in an effort to most accurately stage patients and, thereby, design therapy. Although this procedure is safe, it is frequently uncomfortable for the patient and a source of anxiety and expense.1,2 Lim and colleagues report a retrospective analysis of 192 patients treated at the National Cancer Centre of Singapore. These patients were found to have stage I or II disease after clinical and CT scanning before bone marrow analysis. Clinical data were catalogued, including age, sex, presence of B symptoms, white blood count (WBC), platelet count, hemoglobin, serum lactate dehydrogenase (LDH), serum 2-microglobulin (B2M) , presence of extranodal disease, presence of bulky disease (defined as tumor diameter of > 7 cm) and radiological stage. A low hemoglobin was defined as < 10 g/dL; a low WBC as < 4 × 109/L, and a low platelet count as < 100 × 109/L. Bone marrow was considered involved if either the aspirate or trephine biopsy showed the presence of lymphomatous infiltrate by standard morphological assessment.

Overall, bone marrow involvement was found in 7 patients (3.6%). Comparing the clinical measures between patients with and without marrow involvement, a low hemoglobin (P = 0.02), low WBC (P = 0.007) and bulky disease (P = 0.006) were associated with bone marrow involvement. Among the 120 patients without bulky disease, a hemoglobin level of more than 10 g/dL and a WBC of > 4 × 109/L, only one patient had bone marrow involvement (0.83%; 95% CI, 0.02%-4.6%). The presence of any of the 3 risk factors (low hemoglobin, low WBC, or bulky disease) gave a positive predictive value of 8.3% for bone marrow involvement whereas the absence of all 3 factors gave a negative predictive value of 99.2% for bone marrow involvement.

Comment by William B. Ershler, MD

This was a careful retrospective analysis with some practical findings for the clinician. However, currently patients undergoing staging for lymphoma will have marrow aspirate sent for flow cytometry or even for PCR analysis looking for rearrangement of the immunoglobulin gene. These measures are likely to be more sensitive than routine histological examination alone, and thus might have yielded a higher than 3.5% marrow involvement in this same series.

That stated, outside of a research setting the need to be absolutely certain whether bone marrow is involved is germane only if treatment hinged on the results. In this regard the recent follow-up from the Southwest Oncology Group (SWOG) trial in which 401 patients with limited stage aggressive lymphoma were randomized to either 8 cycles of CHOP chemotherapy or 3 cycles followed by involved field radiation is of importance. Although the early analysis3 showed an advantage for those randomized to the combined modality treatment, the long-term follow up demonstrated an increase in late relapses and lymphoma-related deaths in patients receiving short-course chemotherapy and involved field radiation.4 One might conclude from this study that even patients with early stage diffuse B-cell lymphoma should be treated with chemotherapy alone, and thus bone marrow aspirate would be a non-essential component of the pretreatment evaluation. Clinicians are keen on having as much information as possible before starting treatment. However, in balance, the avoidance of the significant anxiety and pain associated with the marrow procedure may over shadow the importance of the findings (positive or negative) in patients for whom the treatment will likely be the same under either circumstance. The study from the Singapore Cancer Centre is reassuring that for selected early stage patients (those with normal blood counts and non-bulky disease), less than 1% would have been positive if the procedure were performed.


1. Bain BJ, et al. BMJ. 2003;121:949-951

2. Vanhelleputte P, et al. J Pain Symptom Manage. 2003;26:860-886.

3. Miller T, et al. N Engl J Med. 1998;339:21-26.

4. Miller TP, et al. Blood. 2001;98:724a.

William B. Ershler, MD, INOVA Fairfax Hospital Cancer Center, Fairfax, VA; Director, Institute for Advanced Studies in Aging, Washington, DC, is Editor for Clinical Oncology Alert.