Indian Journal of Cancer Home 

[Download PDF]
Year : 2014  |  Volume : 51  |  Issue : 1  |  Page : 30--34

Non-haematopoietic malignancies metastasing to the bone marrow: A 5 year record-based descriptive study from a tertiary care centre in South India

P Mishra, S Das, R Kar, SE Jacob, D Basu 
 Department of Pathology, Jawaharlal Institute of Postgraduate Medical Education and Research, Pondicherry, India

Correspondence Address:
R Kar
Department of Pathology, Jawaharlal Institute of Postgraduate Medical Education and Research, Pondicherry


Introduction: Bone marrow involvement by a non-haematological malignancy gives an opportunity to identify the lodgement, invasion of metastatic cells and the response of the host to the tumor cells. The study was undertaken to assess the involvement of bone marrow with non-haematopoietic malignancies and its significance in establishing primary diagnosis in clinically unsuspected cases. Materials and Methods: This was a descriptive study which included record review of the departmental archives for the last five years (January 2007 to December 2011). Eighty four cases were studied; which included clinically diagnosed non-haematological malignancy for staging or symptomatic cytopenias/bony lesions (group 1, n = 63), clinically suspected bone marrow metastasis of unknown primary malignancy due to symptomatic cytopenias/bony lesions (group 2, n = 07) and clinically unsuspected malignancy with incidentally detected bone marrow metastasis (group 3, n = 14). Results: Bone marrow metastases of solid tumors were identified in 23 cases (27.3%) which included 9 cases from group 1, 14 cases from group 3 and nil in group 2. Of the 14 cases in group 3, in 12 cases a definitive diagnosis could be made by correlating clinicoradiological findings with morphology and immunohistochemistry. The most common tumor in pediatric cases were neuroblastoma and Ewing«SQ»s sarcoma (40%) and in adult«SQ»s adenocarcinoma of gastrointestinal tract (30.7%) was the commonest. Conclusion: Bone marrow metastasis can masquerade as a primary haematopoietic disorder; however its detection has both therapeutic and prognostic significance. Immunohistochemistry is a useful adjunct to morphology in reaching a definitive diagnosis.

How to cite this article:
Mishra P, Das S, Kar R, Jacob S E, Basu D. Non-haematopoietic malignancies metastasing to the bone marrow: A 5 year record-based descriptive study from a tertiary care centre in South India.Indian J Cancer 2014;51:30-34

How to cite this URL:
Mishra P, Das S, Kar R, Jacob S E, Basu D. Non-haematopoietic malignancies metastasing to the bone marrow: A 5 year record-based descriptive study from a tertiary care centre in South India. Indian J Cancer [serial online] 2014 [cited 2022 Aug 15 ];51:30-34
Available from:

Full Text


Metastasis of non-haematopoietic tumor cells to bone marrow was reported as early as in 1834, [1] but a collected series of such cases was not published until 1936. Rohr and Hegglin, [2] studied sternal aspirates of 75 patients with cancer and found marrow involvement in 11 of 13 patients with bone metastases. In 1958, McFarland and Dameshek [3] described a simplified technique for bone marrow biopsy, and the ensuing experience demonstrated that one could discover unsuspected malignant disease and in many cases confirm the finding obtained by aspiration. Bone marrow involvement by haematological malignancies is commonly found; however, specially diagnosing a non-haematological malignancy from marrow is a rare entity. Study of bone marrow not only gives an opportunity to identify the lodgement, invasion and extra vascular spread of metastatic cells at a cellular level; it also gives information regarding the reactions of the host marrow. Detection of metastatic tumors in the bone marrow is crucial for clinical staging, detecting response to treatment and the overall survival. Recognition of metastasis in random biopsies presents challenges to pathologists when diagnosing the primary focus. [4] This study was undertaken to comprehensively analyze bone marrow metastasis in non-haematological malignancies diagnosed at a single tertiary care centre in South India over the five years.

 Materials and Methods

This was a descriptive cross-sectional study conducted in the department of Pathology which included record review of the departmental archives for the last five years from January 2007 to December 2011. A total of 2426 bone marrow aspirations most of which also had biopsies, were done during this period for the diagnosis of various haematological and non-haematological disorders. Of these 2426 cases, there were 84 cases of non-haematological malignancies in which bone marrow study was done. These 84 cases were selected for further study and were grouped as under:

Group 1: Clinically diagnosed non-haematological malignancy for staging or those who had symptomatic cytopenias/bony lesions (n = 63).

Group 2: Clinically suspected bone marrow metastasis of unknown primary malignancy due to symptomatic cytopenias/bony lesions (n = 07).

Group 3: Clinically unsuspected malignancy with incidentally detected bone marrow metastasis (n = 14).

Bone marrow aspirate and biopsies were performed from the posterior iliac spine. Bone marrow aspiration, imprint smears and peripheral smears were stained by Romanowsky stains, either by Giemsa or Leishman. Trephine biopsies after standard processing were stained with Haematoxylin and Eosin. Special stains like Reticulin, Periodic Acid Schiff (PAS), Mucicarmine (MC) and Alcian blue-PAS were done wherever necessary. Morphological features like arrangement of tumor cells, presence or absence of special features like rosettes, pleomorphisim of tumor cells, and marrow response to the tumor cell in the form of marrow necrosis, fibrosis and osteomyelosclerosis were studied. Tumor burden was assessed using the criteria put forth by Frisch et al.[5] Immunohistochemistry (IHC) using standard technique was used wherever required and the panel of markers were chosen based on morphology and clinical history. Standard descriptive statistics was used.


The total number of positive cases showing marrow metastasis was 23 which included 9 from group 1and 14 from group 3 and none in group 2 [Table 1]. The marrow aspiration, biopsy slides and the case records of these 23 cases were studied in detail. Among the bone marrow positive cases for metastasis (n = 23), there were 15 males and 8 females; 10 were in paediatric age group (<14 years), and 13 were in adult age group. Lymphadenopathy was seen in 6 (26%) cases and 4 (17.3%) cases had hepatosplenomegaly. Peripheral smear showed leucoerythroblastic picture in 6 (26%) cases.{Table 1}

Indication for bone marrow

Group 1: Constituted 63 cases of known malignancy the clinical diagnosis of which are summarised in [Table 2]. Out of these 63 cases, primary indication for bone marrow examination was staging in the majority (54 cases). In the remaining cases marrow was done due to unexplained haematological abnormalities like leucoerythroblastic anaemia, cytopenias (5 cases) and bony lesions detected radio graphically like lytic or sclerotic lesions (4 cases). Among these 9 cases showed marrow metastasis.{Table 2}

Group 2: These seven cases were worked up for skeletal lytic lesions with a clinical suspicion of bone metastasis. However, no case was positive in this group.

Group 3: There were 14 cases in this group where bone marrow metastasis was clinically unsuspected but incidentally detected in all the cases. The bone marrow was performed primary for unexplained cytopenias and/or bony lesions. Morphologic features coupled with IHC, detailed clinical and radiological review helped in ascertaining the primary site in 12 of these 14 cases.

The salient clinico-haematological profile of the 23 bone marrow metastasis positive cases is summarized in [Table 3].{Table 3}

Radiological findings

Of the 84 cases that were studied only in 17 cases there was a radiological suspicion of marrow involvement by non haematological malignancy, out of which 7 cases showed metastatic deposits in bone marrow examination (positive predictive value - 41.2%).

Bone marrow aspirate findings

Bone marrow aspiration was done in all the 84 cases and biopsies were done in only 67 cases. The 17 cases where biopsies were not done mostly belonged to the pediatric age group. Bone marrow metastasis was found in 23 out of 84 (27.3%) cases. In 19 cases, bone marrow aspiration showed metastatic deposits, in two cases tumor was picked up in the imprint smear and in two cases both aspirate and imprint smears were negative; however, biopsy was positive. Of the 23 positive cases, in 18 cases bone marrow biopsy was done and all showed infiltration by malignant cells.

Of the 21 positive cases in bone marrow aspirate, 12 cases showed small round cell morphology, five of them in addition showed features like rosettes and neuropil [Figure 1]a and b in the background. Five cases showed features of adenocarcinoma, one of which showed signet ring cell morphology. In the remaining four cases; cells exhibited significant degree of pleomorphisim.{Figure 1}

Most of the non-haematopoietic malignancies in children metastasing to bone marrow were small round cell tumors commonest being neuroblastoma and Ewing's sarcoma each comprising four cases, rhabdomyosarcoma in one case and a single case reported as metastatic small round cell tumor which could not be further sub classified. Of the 13 cases in adults, eight cases were diagnosed as adenocarcinoma, with primaries being located in GI tract in four; one in stomach and three from colon. There were also two cases of lung adenocarcinoma, one case from breast and in one case the primary site of origin could not be ascertained. The remaining five cases composed of two cases of Ewing's sarcoma and one case each of neuroblastoma, clear cell sarcoma and choriocarcinoma.

Bone marrow biopsy findings

In the bone marrow biopsies, eosinophilia was seen in two cases and reactive plasmacytosis was seen in one case as part of tumor myelopathy. Stromal changes like fibrosis of grade 3 were seen in four cases. Osteomyelosclerosis was seen in four cases. Necrosis was seen in two cases. Assessment of tumor burden in bone marrow biopsy was done in 18 cases and categorized based on the classification as proposed by Frisch et al.[5] There was one case in category A, seven cases in category C, ten cases in category D and nil in category B.


The fourteen cases in group 3 were further worked up for the primary tumor with the aid of clinical and radiological findings coupled with an immunohistochemical panel. Immunohistochemistry was done using the markers Cytokeratin (CK)7, CK8, CK20, LCA, PLAP, CD99, CA125, ER, PR, carcinoembryonic antigen (CEA), desmin and synaptophysin depending on the morphology. In 12 cases a definitive opinion was possible. In the final diagnosis we had one case each of Ewing's sarcoma, clear cell sarcoma, rhabdomyosarcoma, choriocarcinoma [Figure 1]c, two cases of neuroblastoma, one case of metastatic adenocarcinoma from stomach and three cases from colon [Figure 2]a and b, and one case each of adenosquamous [Figure 2]c and d and adenocarcinoma from lung. In all these cases diagnosis was confirmed with the primary site of tumor. In two cases, the primary site could not be ascertained. These were reported as metastatic small round cell tumor and metastatic adenocarcinoma respectively. Clinical findings and immunohistochemistry of the above cases are summarized in [Table 4].{Figure 2}{Table 4}


Of the 23 positive cases 6 (26%) had leucoerythroblastic anaemia with immature granulocytes and nucleated red cells in the peripheral smear. Contreras et al.[6] found leucoerythroblastic reaction in 22% of their cases and Leland and McPherson [7] in 33% of their cases. In our study, nucleated red cell and immature granulocytes were not seen in patients in whom the haemoglobin was 10% gm. or more and usually not until the haemoglobin was reduced below 7.5% gm. [Table 3]. Presence of leucoerythoblastic anaemia correlates with the degree of reactive bone marrow fibrosis than with the extent of malignant infiltration. [8] In our study 4 out of these 6 cases showed marrow fibrosis.

Bone marrow aspiration was done in 84 patients. Twenty three cases showed metastatic deposits, (27.3%) which is comparable to the previous studies in India. [9] Twenty one cases were picked up in aspirates, one aspirate was diluted and one was reported as normal active but biopsy showed metastatic deposits. Bone marrow aspiration showed normal or increased cellularity in 47.8% cases while in 52.2% cases normal hemopoietic elements were reduced. Presence of small round cells, neuropil in the background, clustering and glandular arrangement of the cells were helpful findings to clinch to a definitive diagnosis.

In our study the most common tumor in children that metastasized to marrow was neuroblastoma and Ewing's sarcoma each being 40% and in adults was adenocarcinoma of GI tract 30.7%. Similar findings were observed by Ozkalemkas et al.,[10] where the most common histologic subtype was adenocarcinoma gastrointestinal tract 42%. However, a study conducted by Mohanty et al.,[11] showed 100% cases as neuroblastoma in pediatric population and in adult population prostatic adenocarcinoma was the most common malignancy metastasing to bone marrow constituting 48% where as adenocarcinoma gastrointestinal tract was 9%.

There have been several previous studies pointing to superiority of the bone marrow biopsy over aspirate smears in diagnosis of metastatic tumor [12],[13] as deposits in the marrow are focal and may often elicit a fibrotic response and therefore aspirates may be negative. Multiple sections of the biopsy enable a much larger volume of the marrow to be studied and allow infiltration to be recognized. [14] Singh et al.[12] reported that bone marrow biopsy was superior to aspiration in finding carcinoma (97% vs. 72%). Similar findings were observed in our study where biopsy vs. aspiration positivity is (100% vs. 91.3 %). However, a study conducted by Sharma et al.,[15] showed an occasional case where metastasis was detected only in aspiration. Bearden et al.,[16] reported that bone marrow aspirates and biopsies were complementary in diagnosis of various solid tumors.

We had more numbers of positive cases in group 3, i.e., clinically unsuspected than in group 1, i.e., staging bone marrow. This may be attributed to the incomplete clinical workup of these cases at the time of presentation. Usually these patients presented at an advanced stage with symptomatic cytopenias or bony lesions which was misconstrued as a haematological malignancy at the outset.

Tumor burden was assessed in bone marrow biopsy and classified into: [5] A- Microcolonies of single cells or clusters, B- Multiple small tumor foci, C- One or more large masses with preserved haematopoietic marrow and D- Total marrow replacement by tumor. In our study we had one case in category A, seven cases in category C, ten cases in category D and nil in category B. Tumor myelopathy like eosinophilia, plasmacytosis was seen in three cases and 10 cases were associated with stromal changes like necrosis, fibrosis and osteomyelosclerosis. According to Mohanty et al.,[11] a number of associated features were observed like osseous metaplasia, desmoplastic changes, necro inflammatory reaction, granulomas and infections which may help to suggest bone marrow metastasis, in the absence of tumor cells in the bone marrow.

Immunohistochemistry was used as an adjunct to the morphological findings in the marrow in elucidating the primary site especially in clinically unsuspected cases. Depending on the morphology of the tumor appropriate immunohistochemistry panel was used to narrow down the differentials as summarized below:

Undifferentiated carcinoma: Pan CK, EMA, Chromogranin, LCA, S100

Pleomorphic sarcoma: Vimentin, Desmin, SMA, CD117, S100, CD99, CD68

Round cell tumor: Pan CK, Vimentin, Desmin, CD99, NSE, LCA, Chromogranin

At times immunohistochemistry can be non-contributory owing to extensive necrosis with lack of viable tissue and sections with scanty tissue. In our study in one case immunohistochemistry was not much of help because of extensive necrosis.


Detection of metastasis in the bone marrow has both therapeutic and prognostic significance. Clinical history, radiological findings, morphology and immunohistochemistry with a panel of antibodies is useful to arrive at a definitive diagnosis. In addition immunohistochemistry has a special role to play in biopsies especially where the nature of the primary tumor is unknown.


1Samson M. Fait remarkable de diathese cancereuse. Gaz Med de Paris 1834;2:140.
2Rohr K, Hegglin R. Tumorzellen in sternapunktat. Deutsch Arch F Klin Med 1936;179:61-79.
3Mc Farland W, Dameshek W. Biopsy of bone marrow with the Vim Silverman needle. JAMA 1958;166:1464-8.
4Papac RJ. Bone marrow metastasis. Cancer 1994;74:2403-13.
5Frisch B, Bartl R, Mahl G, Burkhardt R. Scope and value of bone marrow biopsies in metastatic cancer. Invasion Metastasis 1984;4:12-30.
6Contreras E, Ellis LD, Lee RE. Value of the bone marrow biopsy in the diagnosis of metastatic carcinoma. Cancer 1972;29:778-83.
7Leland J, MacPherson B. Hematologic findings in cases of mammary cancer metastatic to bone marrow. Am J Clin Pathol 1979;71:31-5.
8Rubins JR. The role of myelofibrosis in malignant myelosclerosis. Cancer 1983;51:308-11.
9Mehdi SR, Bhatt MLB. Metastasis of solid tumors in bone marrow: A study from Northern India. Indian J Hematol Blood Transfus 2011;27:93-5.
10Ozkalemkas F, Ali R, Ozkocaman V, Ozcelik T, Ozan U, Ozturk H, et al. The bone marrow aspirate and biopsy in the diagnosis of unsuspected non-hematologic malignancy: A clinical study of 19 cases. BMC Cancer 2005;5:144-52.
11Mohanty SK, Dash S. Bone marrow metastasis in solid tumors. Indian J Pathol Microbiol 2003;46:613-6.
12Singh G, Krause JR, Breitfeld V. Bone marrow examination for metastatic tumor aspirate and biopsy. Cancer 1977;40:2317-21.
13Ingle JN, Tormey DC, Tan HK. The bone marrow examination in breast cancer; diagnostic considerations and clinical usefulness. Cancer 1978;41:670-4.
14Basu D, Singh T, Shinghal RN. Micrometastasis in bone marrow in breast cancer. Indian J Pathol Microbiol 1994;38:159- 64.
15Sharma S, Murari M. Bone marrow involvement by metastatic solid tumors. Indian J Pathol Microbiol 2003;46:382-4.
16Bearden JD, Ratkin GA, Coltman CA. Comparison of the diagnostic value of bone marrow biopsy and bone marrow aspiration in neoplastic disease. J Clin Pathol 1974;27:738-40.