Granulocyte Colony Stimulating Factor (G-CSF)

Neutropenia is a dose limiting toxicity of chemotherapy. It results in delay and dose reduction both of which adversely affect outcomes of treatment. Myeloid growth factors are biological agents that stimulate the production go granulocytes and offset the myelosupressive effect of chemotherapy. Two myeloid growth factors are available Granulocytic colony stimulating factor (G-CSF) and granulocytic monocytic colony simulating factor (GM-CSF). This article will discuss G-CSF as it is used more often than GM-CSF. Commertially available G-CSF is made by recombinant DNA technology and may be produced in E. coli (Filgrastim) or chinese hamster ovary cell lines (lenograstim). The half life of filgrastim can be increased by covalently linking it to polyethylene glycol (PEG) and converting it to pegfilgrastim.

Mechanism of Action of G-CSF

G-CSF is a 174 amino acid peptide the gene for which is on chromosome 17. It has a molecular weight of 18kDa. It is produced by monocytes, macrophages, fibroblasts, endothelial cells and keratinocytes in response to inflammatory cytokines and bacterial endotoxin.

G-CSF acts via the G-CSF receptor. G-CSF receptor is a transmembrane receptor that form a homodimer on binding G-CSF. Activation of G-CSF receptor results in activations of  JAK/STAT, SRC family of kinases, PI3/AKT and Ras/ERK 1/2. The details of the pathway are not completely understood.


Activations of G-CSF has the following effects that lead to increased production of neutrophils

  1. Increased Proliferation of Neutrophilic precursors
  2. Shortened neutrophilic precursor bone marrow transit time
  3. Functions maturation of neutrophils – increased chemotaxis, phagocytosis and antibody dependent cytotoxicity

G-CSFs Available for Clinical Use

G-CSFs for clinical use is manufactured by recombinant DNA technology. Two molecules are available for clinical use. Filgrastim is produced using E. coli and lenograstim is obtained from Chinese hamster ovarian cells. Lenograstim is glycosylated (4% glycosylation).

Filgrastim on subcutaneous administration filgrastim has a half life of 2.5-5.8 hours. The drug is eliminated by uptake be G-CSF receptors on neutrophils and glomerular filtration. Pegylation, that involves attaching a 20kDa polyethylene glycol (PEG) molecule to the N terminal eliminates renal elimination prolonging the half life to 27-47 hours. The product, pegfilgrastim, is only eliminated by binding to neutrophil G-CSF receptors, patients with low neutrophil counts have a lower clearance. Prevention of glomerular filtration allows administration of pegfilgrastim only once in a chemotherapy cycle.

Indications for G-CSF

The discussion that follows applies to filgrastim and perfilgrastim as these drugs are used more commonly than lenograstim. The general principle apply to lenograstim but readers are advised to refer to information on lenograstim for details of use and adverse effects.

  1. Primary prevention of febrile neutropenia (FN) in patients with non-Myeloid malignancy on chemotherapy: Patients where on chemotherapy protocols that have a risk of febrile neutropenia equal to or greater than 20% should be administered G-CSF.
  2. Prevention of recurrence of febrile neutropenia: G-CFS may be used to prevent recurrence of febrile neutropenia in patients who have had an episode of infection in a previous chemotherapy cycle.
  3. Treatment of patients with febrile neutropenia: Initiating therapy with G-CSF after febrile neutropenia has set in has not been shown to decrease mortality of antibiotic use. It may however be used in patients who are at high risk of mortality.
  4. Mobilisation of stem cells for stem cell transplant
  5. Use in patients with myeloid malignancies: There is an apprehension that G-CSF may stimulate leukaemia cells and G-CSF is not used in induction. It may however be used after induction to reduct the duration of neutropenia.

Filgrastim is administered in a dose of 5μg/kg/day subcutaneously, by a short iv infusion or prolonged intravenous infection. therapy should be initiated at least 24 hours after the  chemotherapy. The adult dose of pegfilgrastim is 6mg. The paediatric dose depends on the weight of the child. Children less than 10 kg: 0.1 mg/kg, those between 10 to 20 kg be administered 1.5 mg, between 21 to 30 kg be administered 2.5 mg and between 31 to 44 kg administered 4 mg. Children weighing 45kg or more should be administered the adult dose of 6 mg. Pegfilgrastim should not be administered less than 14 days after a cycle of chemotherapy. It should be administered more than 24 hours after a cycle of chemotherapy.

Adverse Effects

  1. Bone Pain: Bone pain is the commonest side effect with about 20-30% of the patients suffering the side effect.
  2. Rare but serous side effects include splenic rupture, acute respiratory distress syndrome,  precipitation of sickle cell crisis and capillary leak syndrome

Clinical Features of Megaloblastic Anaemia

Megaloblastic anaemia is a macrocytic anaemia resulting from the deficiency of vitamin B12 or folic acid characterised by the presence of megaloblasts in the bone marrow. It has haematological and neurological manifestations. The haematological manifestations are seen with folate as well as vitamin B12 deficiency. Folate deficiency in adults does not affect the nervous system.

Cobalamin deficiency is slow and “pure”. Folate deficiency is rapid and “impure”. Deficiecy of vitamin B12 occurs because of loss of intrinsic factor resulting in an isolated defect of B12 absorption. No other nutrients are affected. The body stores of B12 can last months. This results in B12 deficiency being a slow and “pure” deficiency. Symptoms come on slowly, over months. Folate deficiency evolves relatively quickly and is most commonly because of alcoholism or malabsorption. It is associated with other deficiencies and is rapid and “not pure”.


Manifestationf o megaloblastic anaemia

Figure 1. Clinical Manifestations of Megaloblastic Anaemia

Haematological Manifestations

Haematological changes resulting from vitamin B12 deficiency and folate deficiency are indistinguishable. Megaloblastic anaemias are macrocytic anaemia but macrocytosis is not specific to megaloblastic anaemia. It is however exceptional for other diseases characterised by macrocytosis to have an mean capsular volume (MCV) > 110fl.  This value can considered the threshold above which an anaemia is unlikely to be anything other than megaloblastic anaemia.

The earliest change in a megaloblastic anaemia is macrocytosis. This precedes changes in erythrocyte indices. Changes in mean capsular haemoglobin (MCH) follow and then the MCV rises. Haemoglobin usually falls after the MCV increases to >97 fl. As the severity of anaemia increases the peripheral smear shows aniscytosis and poikilocytosis, nucleated cells, Howell-Jolly bodies and Cabot’s ring. Microcytes and erythrocyte fragments that represent dyserythropoiesis may be seen. Polychromasia is absent and this distinguishes megaloblastic anaemia from haemolytic anaemia.

The term megaloblatic anaemia is a misnomer. The disease is actually a panmyelosis.  Erythroid, myeloid and megakaryocytic series are affected. Thrombocytopenia and leucopenia (neutropenia and to a lesser extent lymphopenia) usually occur late in the course. It is uncommon for patients with mild anaemia to have platelets and neutrophils but occasionally changes in leucocytes and/or platelets may dominate.

Iron deficiency or β-thalassaemia trait result in microcytosis and hypochromia and may incidentally co-exist with megaloblastic anaemia. Co-existence of either of these diseases with megaloblastic anaemia may mask macrocytosis of megaloblastic anaemia. Presence of hypersegmented neutrophils in a patients with normocytic normochromic anaemia should raise the suspicion of a megaloblastic anaemia co-existing with Iron deficiency or β-thalassaemia trait.

Neurological Manifestations

Cobalamine deficiceny causes neurological dysfunction. Folate deficiency causes symptoms only in children. Children with inborn errors of folate metabolism may have myelopathy, brain dysfunction and seizures.

The neurological manifestations of B12 deficiency are a result of a combination of upper motor neuron manifestations from subacute combined degeneration of the spinal cord, sensory and lower motor neuron manifestations from peripheral neuropathy and neurophychiatratic manifestations. Subacute combined degeneration of the spinal cord (SACD) is a degerative disease of the spinal cord involving the posterior and lateral column (corticospinal and spinoceribellar tracts) that starts in the cervical and the thoracic region.

The earliest neurological manifestations are impaired sense of vibration and position and symmetric dysesthasia that involve the lower limb. This is frequently associated with sensory ataxia. With progression spastic paraparesis develops. The patients have brisk knee reflexes, reflecting an upper motor neuron involvement and depressed ankle reflex, reflecting a peripheral neuropathy. Bladder involvement is unusual. Some patients may have optic atrophy.

Neuropsychiatric manifestation include memory loss, depression, hypomania, paranoid psychosis with auditory and visual hallucinations.

Other manifestations

Skin and nails can show pigmentations. Mucosa of the villi undergoes megalobkastic change resulting in temporary malabsorption.

Response to therapy

Haematological Recovery

  • Day 1: Feeling better
  • Day2-3: Reticulocytosis appears
  • Day 7-10: Peak retuculocytosis
  • Day 15 onwards: Neutrophilic hypersegmentation disappears
  • Day 56 (8 weeks): Blood counts become fully .normal

Neurological Recovery

Neurologic improvement begins within the first week also and is typically complete in 6 weeks to 3 months. Its course is not as predictable as hematologic response and may not be complete.



The M-Band

Monoclonal Gammopathy-02

Figure 1. Each plasma cell produces a different type of antibody. Normal γ globin band is depicted in the left column. The plasma cell numbers are normal and each produces an antibody with a different amino acid structure and electrophoretic mobility. Patients with monoclonal gammopathy have expansion (increase number) of a plasma cell clone (red in the diagram) resulting in the production of a disproportionate large amount of immunoglobulin from one type of plasma cell. This results in the M Band (see below). Patients with polyclonal gammopathy have an expansion (increased number) of plasma cells. This is usually occurs in response to infection/inflammation that result in production of a diversity of antibodies. The diversity is reflected in increase in the γ but as no one clone dominates the sharp M band is not seen.

What is an M-Band?

Immunoglobulins are antigen binding molecules secreted by plasma cells. Immunoglobulins bind antigens and play a role acquired immunity. Plasma cells develop from antigen exposed B-lymphocytes. The process of maturation of lymphocytes involves inducing mutations in region of the immunoglobulin gene that encodes for antigen binding regions, the hypervariable regions. This makes the DNA and consequently the amino acid sequence of the immunoglobulin secereted by a plasma cell unique. This is true even when two plasma cells make antibody against the same antigen or antigenic epitope (see figure 1).

Monoclonal Gammopathy-01

Figure 2. The serum protein separate into many bands on electrophoresis. The albumin is a dark band closest to the anode. This is followed by the α1, α2, β and γ bands. The immunoglobulin are mainly found in the γ globulin band but some may be found in the β globin band. The electrophoretic mobility of a molecule depends on the charge it carries which in turn depends on the amino acid sequence. Amino acid sequence determines the antigen specificity and differs between antibodies resulting in a slight variation in electrophoretic mobility of immunoglobulins and resulting in the γ region being a broad band.

The amino acid sequence determines the charge on the immunoglobulin. The electrophoretic mobility is determined by the charge. Majority of the immunoglobulins move to the γ-globulin fraction of serum proteins, some move with β-globulin. The γ-globulin band is a wide electrophoretic band reflecting the diversity in electrophoretic mobility of immunoglobulins arising from the diversity in amino acid sequences (figure 2).

Monoclonal Gammopathy-03

Figure 3. Patinets of monoclonal gammopathies have an expansion of one clone of plasma cells. This reflects in production of a disproportionally large amount of immunoglobulin with identical electrophoretic mobility resulting in a dense band with in γ globin region

Patients of monoclonal gammopathies have clonal expansion of plasma cells. The cells of a clone have identical DNA and produce identical immunoglobulin molecules. When the clone grows to level that it forms a significant proportion of the plasma cell pool the immunoglobulin it produces forms a significant proportion of the total serum immunoglobulins. The identical electrophoretic mobility of molecules produced by the clone results in a disproportionately large number of immunoglobulin concentrating to a point on electrophoresis forming a band.  This is known as the M band.  Lymphoma cells, notably those of lymphoplasmacytic lymphoma, can secrete immunoglobulin and are associated with an M band for similar reasons.

Diseases associated with an M-Band

The M-Band is a serum marker for plasma cell dycrasias and Waldenström macroglobulinemia. IgM and non-IgM (mainly IgG and IgA) monoclonal bands have differing clinical implications. The former is more commonly associated with lymphoproliferative disease and the latter with plasma cell dycrasias. The presence of an M band only indicates a clonal expansion of immunoglobulin producing cells. It does not indicate malignancy. The diagnosis of malignancy is made by features that suggest end organ damage. The absence of end organ damage indicates a premalignant disease including monoclonal gammopathy of uncertain significance (MGUS), soldering multiple myeloma or smoldering Waldenström macroglobulinemia.  The evidence of end-organ damage includes

  1. non-IgM Monoclonal Gammoathies: CRAB (elevated calcium, renal involvement, anaemia and osteolytic (bone) lesions) creatinine,
  2. IgM Monoclonal Gammapathies: Anemia, constitutional symptoms, hyperviscosity, lymphadenopathy, or hepatosplenomegaly that can be attributed to the underlying lymphoproliferative disorder if diagnosis is Waldenström macroglobulinemia or CRAB (elevated calcium, renal involvement, anaemia and osteolytic (bone) lesions) creatinine if the diagnosis of IgM myeloma

False positive M-Band

The presence of M band indicates presence of a clonal expansion of plasma cells. When end organ damage co-exists with M band a diagnosis of a malignancy (multiple myeloma or Waldenström macroglobulinemia) is made. In the absence of end organ damage the diagnosis of a premalignant disease is made. Proliferation a of plasma cells are seen in infections/inflammation. These are polyclonal and result in s polyclonal gammopath. They do not result in the presence of an M-band.



Evolution and Spread of HbS

The gene for β globin (OMIM  is present on chromosome 11 (11p15.4) along with other globin genes (ε, γ, γ and δ). This is known as the β-globin cluster . Individuals carrying identical genes on the β-globin gene cluster may not have identical DNA sequences in non-codeing regions of the DNA of the cluster. The non-coding regions include segments of DNA between genes and introns within genes. . Differences in DNA exist between individuals every 1000-2000 bases in the form of single nucleotide polymorphisms (SNPs). Single nucleotide polymorphisms are variations in a single nucleotide that occurs at a specific position in the genome. Many of these differences have no consequences on gene expression because either they do not result in change in amino acid sequence or they occur in regions of DNA that neither code for the gene nor regulate the gene. SNPs evolve by spontaneous mutations over time. The lesser the number of such differences between two individuals closer the individuals are the each other genetically (and in terms of evolution). Fewer differences in SNPs between individuals mean a more recent common ancestor.

One of the meanings of the word haplotype is a pattern of SNPs. A haplotype may be considered as a DNA “environment” in which the gene(s) occurs. This “environment” is created by the sequence of single nucleotide polymorphisms in which the gene(s) exists. As mentioned above differences in SNPs (and hence the “environment” the gene(s) exist in) evolve by spontaneous mutations over period of time. Fewer the differences between the “environments” the genes occurs in the more the likelihood that they come from related individuals.

HbS results from a single base substitution in the codon 6 of the β-globin gene. GAG becomes GTA resulting in substitution of valine for glutamate. This change results in a haemoglobin that crystallizes in hypoxic conditions resulting in a haemolytic anaemia. HbS occurs in diverse population groups including African, Mediterranean, Middle-Eastern and Indian. Is the haplotype of the HbS gene in these regions similar?

The HbS mutation occurs on five different haplotypes four African and one Arab-Indian. The mutation is the same (GAG to GTA on codon 6) but the SNPs are different. The haplotypes are

  1. Senegal: The Senegal HbS haplotype is found in Atlantic West Africa and Portugal
  2. Benin: The Benin HbS haplotype is found Central West Africa, Northern Africa and Mediterranean Europe (Greece, Sicily)
  3. Central African Republic or Bantu: The Central African Republic or Bantu is found in South Central and Eastern Africa
  4. Cameroon: The Cameroon haplotype is found in the Eton ethnic group of eastern Cameroon
  5. Arab-Indian: The Arab-Indian haplotype is the only non-African phenotype of HbS found in the eastern oasis of Saudi Arabia and India.

Origin of Haplotypes

There are two theories about the origin of haplotypes. The first, and the more accepted one, states that the five haplotypes arose from five independent mutations. An alternative hypothesis states that HbS mutation occurred only once and spread to other haplotypes by gene conversion.


Haplotypes and Severity of Symptoms

Symptoms of sickle cell anaemia are a consequence of crystallisation of haemoglobin under hypoxic conditions. HbF inhibits sickling. Patients with high HbF have fewer symptoms. The Arab-Indian and the Senegal haplotype are associated with higher HbF levels (17% and 12.4% respectively). In general patients carrying these haplotypes have milder symptoms than the Bantu or Benin haplotypes (Blood 2014; 123: 481)


Haplotypes and Human Migrations

Trade, conquests and human migrations (voluntary and slave trade) have disseminated the African haplotypes beyond Africa.

  1. The Mediterranean: Most of the Mediterranean (Greece and Scilly) has the Benin haplotype. This reflects pre-historic migrations from Central West Africa along the then fertile Sahara to North Africa. From here it spread to the Mediterranean via the interactions (Trade and wars) between the two regions. The only exception is Portugal. Portugal has the Senegal haplotype which reflects the trading contacts between Portugal and Atlantic West Africa (Angola and Mozambique).
  2. Americas: Neither the native americans nor the original European settlers to the Americas carried the HbS gene. HbS was imported to the Americas with the slaves from Africa. Jamaica was an important slave import hub and records for where tthe slaves arrived from are available. Jamaica has 73% Benin haplotype, 17% Bantu and 10% Senegal haplotypes. These numbers are close to the actual number of slaves who arrived in Jamaica from regions of Africa where these haplotypes are prevalent. Similarly the distribution of haplotype correspond to the origins of slaves in Baltimore and South Carolina (Mariam Bloom. Understanding Sickle Cell Disease, Page 34).
  3. Arab or Indian: It is not clear if the Arab-Indian haplotype originated in India or Saudi Arabia. But considering that all of tribal India has only one haplotype but the East and West Arabian Peninsula have different haplotypes it is possible that the haplotype originated in India.
  4. Spread to Other Parts: As opposed to the era of slave trade modern migration of people in the recent past have been voluntary. These populations have spread across the world as have those form mediterranean but to a lesser extent. These migrations have introduced the HbS gene in areas where it was not indigenous.


Anaemia with Hyperbilirubinaemia

A 49-year-old female presented with dyspnoea on exertion of 1 month duration. Examination reviled pallor and icterus. There was no lymphadenopathy, clubbing, koilonychia, platonychia, petechiae or purpura. There was no oedema of feet. The pulse was 90/min and the blood pressure 130/70 mm of Hg. Examination of the respiratory, cardiac and nervous systems did not show any abnormality. There was no organomegaly.

The haemoglobin was 4.9 g/dL with an erythrocyte count 1.37 x 1012/L, haematocrit of 16%, MCV of 116.78 fL, MCH of 35.77 pg and MCHC 30.63 of g/L.  The leucocytes count was 2800 with 35% neutrophils and 65% lymphocytes. The platelet count was 90 x 109/L. The peripheral smear showed macrocytosis and anisocytosis. Hypersegmented neutrophils were seen. The reticulocyte count was 3%.

The bilirubin was 2.1 mg/dL with a direct bilirubin of 1.8mg/dL and an indirect bilirubin of 0.3mg/dL. The Lactate dehydrogenase was 1417IU (normal 105 – 333 IU/L).

Anaemia and unconjugated hyperbilirubinaemia are characteristic of haemolysis. Does this patient have haemolytic anaemia?

Haemolysis shortens erythrocyte lifespan and results in increases haemoglobin breakdown. Haemoglobin is made of heme and globin. Heme consists of porphyrin ring at the centre of which is iron in the ferrous state. Iron released from catabolism of heme is reused. The porphyrin ring is catabolised to bilirubin. The bilirubin is transported to the liver for conjugation and excretion (see haemoglobin catabolism). Patients of haemolytic anaemia have unconjugated hyperbilirubinaemia because the increased bilirubin production overwhelms the hepatic bilirubin conjugation capacity.

One of the characteristics of megaloblastic anaemia is ineffective erythropoiesis. Ineffective erythropoiesis is defined as a sub-optimal (fewer) production of mature erythrocytes from a proliferating pool of immature erythroblasts. Each immature erythroblast produces less than the optimal number of erythrocytes because of premature death of erythroid precursors including haemoglobinized precursors. The haemoglobin released from haemoglobinized erythroid precursors is catabolised in the same manner as haemoglobin released from lysed erythrocytes (see haemoglobin catabolism). Megaloblastic anaemias are associated with unconjugated hyperbilirubinaemia because of death of haemoglobinized erythroid precursors.

The treatment of haemolytic anaemia and megaloblastic anaemia are different? How does one differentiate megaloblastic anaemia from that because of haemolytic anaemia? Does this patients have a haemolytic anaemia or megaloblastic anaemia?

Haemolytic anaemia is characterised by shortened erythrocyte survival. Erythrocytes survival is estimated by the use of radionucleotides something that is not possible at most centres. In clinical practice, a shortened erythrocyte survival is inferred from a high reticulocyte count. Reticulocytes are erythrocytes that have been produced in the preceding 24 hours. The erythrocytes survival is about 120 days and about 1% of erythrocytes are produced every day. Consistent with this the normal reticulocyte count is 0.5-1.5%.In patients of haemolytic anaemia, ddestruction of erythrocytes is matched by an increased production by the bone marrow. This manifests as reticulocytosis (see reticulocyte count). Megaloblastic anaemia occurs because of decreased production of erythrocytes and this manifests as reticulocytopenia. The difference between haemolytic anaemia and megaloblastic anaemia is the reticulocytosis in the former reticulocytopenia in the latter. This patient had a high reticulcoyte count but after correction both the reticulocyte production index [0.43] and corrected reticulocyte count [1.07%] were low excluding haemolysis. This patient was evaluated for megaloblastic anaemia.

The haemogram has clues to differentiate between haemolytic anaemia and megaloblastic anaemia. These include

  1. A very high MCV: The MCV is very high. Patients with haemolytic anaemia have a mild elevation in MCV. An MCV value >110fL is almost exclusively found in megaloblastic anaemias because of folate and/or B12 deficiency.
  2. Pancytopenia: B12 and folate deficiency impair DNA synthesis impairing erythrpoieis, myelopoiesis and megakaryopoiesis. Nutritional megaloblastic anaemias because of vitamin B12 and/or folate deficiency may show pancytopenia.
  3. Hypersegmented neutrophils (>5% neutrophils with >5lobes) is a feature of megaloblastic anaemia

Other features of megaloblastic anaemia include rise serum transferrin receptor, increased serum iron, serum ferritin and methemalbumin levels. Like haemolytic anaemia the serum haptoglobin is low and the LDH high. LDH levels in megaloblastic anaemia can ve very high.

This patients had a low serum B12 and was treated with parental B12 (1mg alternate day for 5 doses) and was evaluated for cause of vitamin B12 deficiency. As Schilling’s test was not available a diagnosis of pernicious anaemia was made by documenting gastric atrophy and anti-parietal cell antibodies.

Calreticulin and Myeloproliferative Disease

Myeloproliferative disorders (polycythaemia vera [PV], essential thrombocytosis [ET], progressive myelofibrosis [PMF]) are a group of diseases that are characterised by increased proliferation of blood cells, splenomegaly, myelofibrosis, thrombosis and risk of malignant transformation.  The year 2005 was a landmark year for myeloproliferative diseases. Four groups of scientists identified the presence of JAK2V617F mutations in PV. This mutation is present in about 98% patients with PV. Mutations of exon 12 of the JAK2 gene can be found in 1-2% of the PV. These patients do not show the JAK2V617F mutation. The discovery of these mutations gave a genetic definition PV making diagnosis objective.

PV is diagnosed by the presence primary erythrocytosis in the precession of a JAK2 mutation referred to above. Chronic myeloid leukaemia is diagnosed by demonstrating the BCR-ABL1 translocation. JAK2V617F is also present in 50-60% of ET and PMF. Mutation of the gene MPL is found in 1-2%  patients of ET and 5-10% of the patients with PMF. The presence of these mutation helps make diagnosis. However, The diagnosis of PMF and ET in a large proportion of patients requires exclusion of a reactive disorder and other myeloproliferative diseases because these patients (38-49% of ET and 30-45% of PMF) have no genetic marker.

Two publications have shown that a large proportion of the patients with ET and PMF who do not have JAK have mutation calreticulin (CALR) (N Engl J Med. 2013;369(25):2391-2405,  N Engl J Med. 2013;369(25):2379-2390). In addition to ET and PMF CALR mutations are found in the MDS/MPN overlap disorder and refractory anemia with ring sideroblasts with thrombocytosis (RARS-T). They are rare or absent in other myeloid or lymphoid neoplasms or solid tumors.

Calreticulin (CALR) is a major calcium binding protein. The gene for calreticulin is located on 19p13.2. About a quarter of ET and MF have mutation in the CALR gene. All CALR mutations are localised to exon 9 and generate a 1bp frameshift. As a result of this most or almost all the C terminal negative amino acids and calcium binding sites are lost.  There is a complete loss of the KDEL endoplasmic reticulum binding sequence. These mutations have been identified in the haemopoietic stem cell and progenitor compartments. CALR mutations and JAK2 mutations are mutually exclusive.

CALR mutated myeloproliferative disease have a distinct clinical profile. These patients have a lower haemoglobin, lower leukocyte count, higher platelet count and a lower risk of thrombosis. Patients of PMF carrying a CALR mutation have a longer survival than those carrying JAK2 or MPL mutations. Patients with ET carrying the CALR mutations have a longer survival than those carrying the JALK2 mutation. There is no difference between the survival of ET patients carrying CALR mutations and MPL mutations.

Mutated CALR appears to stimulate STAT pathway. It appears to physically bind with the thrombopoietin receptor to stimulate STAT. The erythropoietin receptor is not needed for this action (Blood. 2015;10.1182/blood-2015-11-681932Blood. 2015;126:LBA-4).




Classification of Lymphoma

Lymphomas are a group of malignancies arising from lymphoid tissue. They have a diverse etiology, pathogenesis, clinical presentation, treatment and outcomes. Morphology alone is insufficient to classify lymphomas but for a long time a pathologist had little other than morphology for diagnosis. By the 1980s many advances that were instrumental in taking lymphoma classification beyond morphology had taken place. These advances included:

  1. Recognition of lymphocyte subtypes, T, B and NK cells and development of immunological and DNA based tests to identify these cells.
  2. Hybridoma technology that made available antibodies which were used initially for lymphoma diagnosis and then in lymphoma treatment
  3. Sanger sequencing made determining the sequence of genes possible
  4. Fluorescent in situ hybridisation (FISH) allowed study the mutations in cells in interphase
  5. Chemotherapy achieved cure in some lymphomas and control in others

These technologies were instrumental in generating information about lymphomas including pathogenesis, genetics, immunophenotype and clinical course. It became apparent that lymphomas are one of the most complex malignancies in terms of pathogeneis diagnosis and treatment. Such is the heterogeneity of lymphomas that one of the aggressive (Burkitts’s lymphoma) and one of the most indolent malignancies (small lymphocytic lymphoma/chronic lymphocytic leukaemia) are both lymphomas.

Historically several lymphoma classifications have came into use. Each specialist looked at lymphomas from a different  and his/her own perspective. To the pathologist it was about defining different histological entities and how these entities related to each other. To the clinician it was about defining entities with distinct treatments and outcomes. To complicate matters similar/same entities were referred to by different names by different groups. The confusion that prevailed highlighted the need for co-operation between experts in the field of lymphoma. The first such attempt of co-operation resulted in  the REAL (Revised European American Lymphoma) classification proposed in 1994 by a group of 19 haematopathologists, the International Lymphoma Study Group. This classification used all available information (including histology, genetics, immunophenotyping and clinical course) to define entities. This approach was adapted by the WHO classifications that followed the REAL classification. The most current classification of lymphomas is the 2008 WHO classification. The milestones in the classification of lymphomas are given in the table below.

Year Classifications Features
1941 Gall and Mallory
  1. First generally accepted classification of lymphoma, defined follicular lymphoma
1947 Jackson Parker
  1. First Classification of Hodgkin Lymphoma
1956 Rapaport (Non-Hodgkin Lymphoma)
  1. Classified lymphomas in to follicular and diffuse and within each category by cell morphology.
  2. Within each category nodular lymphomas had a better outcome.
  3. Continued to regard the origins of large cell lymphomas from non-lymphoid cells
1966 Luke and Buttler
  1. Proposed a classification of Hodgkin lymphoma which from the basis of modern classification.
  2. Recognised nodular sclerosis and mixed cellularity.
  3. Recognized the L&H cell
1974 Kiel Classification (Non-Hodgkin Lymphoma)
  1. Recognised that many lymphomas resemble normal germinal centre.
  2. Classified lymphomas according to lymphocytic differentiation as understood at the time. Suggested the putative normal counterparts of lymphomas.
  3. Classified lymphomas in B and T types
1982 Working Formulation (Non-Hodgkin Lymphoma)
  1. Studied 6 classification schemes in use at the time found none to be superior. Consenseus could not be reached because of lack of agreement between pathologists.
  2. Proposed a formulation to translate amongst schemes.
  3. Stratified outcomes based on outcome of trials conducted in the 1970s. Did not use immunophenotyping.
1994 REAL Classification
  1. Developed by a group of pathologists, international lymphoma study group, that made an attempt to overcome differences and focused on identification of “real” entities by incorporating all (morphology, genetics, immunophenotype and clinical course) knowledge available at the time.
  2. Formed the basis of the currently used WHO classification
2001 and 2008 WHO Classifications
  1. The 2008 WHO lymphoma classification is the current classification
  2. Based on pathology, genetics and clinical outcomes

Classification of Lymphoma

The 2008 WHO classification was a result of international collaboration among pathologists, molecular biologists and clinicians interested in the hematological malignancies. Lymphomas are divided into three groups the

  1. B-cell neoplasm
  2. T and NK cell lymphomas

  3. Hodgkin’s lymphoma.

The non-Hodgkin lymphomas are further divided into into precursor neoplasm and peripheral/mature neoplasm. The peripheral lymphoid tissue have mature lymphocytes (peripheral lymphocytes). The precursor lymphoid cells mature in the bone marrow (B cells) and thymus (T Cells).

Lymphocyte development begins with the lymphoblast. A mature lymphocyte expresses a antigen receptor complex which consists of two parts, the antigen receptor and associated signal proteins. Immunoglobulins serve as antigen receptors of B cells. Immunoglobulins  have a constant and a variable region. The genome has many DNA segments encoding for the variable region. Antibodies have different antigen specificity because different segments are chosen to form the gene of the variable region. A wide array of antibody   specificity (millions) can be generated from combination of these DNA segments. Antibody specificity can be further diversified by a process known as somatic hypermutation referred to below. Cells that are undergoing antibody editing are precursor B cells. B cell maturation occurs when the process of antibody editing is complete. Mature B cells express a complete antigen receptor, IgD and IgM on the surface. Similarly a mature T cell is a cell that has completed the process of editing its T cell receptor.

Precursor Neoplasm

Precursor cells are cells that have not undergone the B or T cell receptor rearrangement. The malignancies of precursor lymphoid tissue incelude T and B cell lymphoblastic lymphomas and acute lymphoblastic leukaemia.

B lymphoblastic lymphoma/leukaemia is further classified into B-lymphoblastic leukaemia/lymphoma with recurrent genetic anomalies and B-lymphoblastic leukaemia/lymphoma that does not show these anomalies (B-lymphoblastic leukaemia/lymphoma NOS). The recurrent anomalies seen in B-lymphoblastic leukaemia/lymphoma are [gene rearrangements]

  1. t(9;22)(q34;q11.2) [BCR-ABL1]
  2. t(v;11q23) [MLL rearranged]
  3. t(12;21)(p13;q22) [TEL-AML1 (ETV6-RUNX1)]
  4. t(5;14)(q31;q32)[IL3-IGH]
  5. t(1;19)(q23;p13.3)[TCF3-PBX1]
  6. hyperdiploidy
  7. hypodiploidy


Neoplasm of the Mature (peripheral) Cells

Neoplasm of mature lymphocytes are classified into B cell neoplasms and T and NK cell neoplasms.


Mature B cell neoplasms

Mature B-cell neoplasm arise from B cells that have undergone B cell receptor rearrangement. Though these cells have their immunoglobulin or T cell receptors rearranged and are referred to as mature the process of maturation is not complete. They undergo a final phase of maturation on exposure to antigens that results in increased antibody avidity. This process takes place in the germinal centre. Antibody avidity is increased by inducing mutations in the DNA segments encoding for the variable regions. This process known as somatic hypermutation.  Somatic hypermutation is a considered to be an evidence of a cell that has passed through the germinal centre (and hence been exposed to antigen). Somatic hypermutations result in a spectrum of avidity (both higher and lower than the original cell). Cells producing highest affinity antibodies survive to form memory B cells or mature to antibody secreting plasma cells. The rest undergo apoptosis. Mutations and apoptosis are two phenomena central to malignant transformation. Germinal centre cells are subject to both. It is not surprising that the germinal centre is the site of the largest number of lymphomas. Diffuses large B Cell lymphoma, follicular lymphoma, Hodgkin’s lymphoma classical and nodular lymphocyte predominant and Burkitts’s lymphoma originate in the germinal centre. Together these constitute almost two third of the lymphomas. Most mantle cell lymphomas originate from cells that have yet to enter the germinal centre. Chronic lymphocytic leukaemia, marginal zone lymphomas, plasma cell neoplasms and lymphoplasmacytic lymphomas arise from cells that have passed through the germinal centre.

Diffuse large B cell lymphoma (DLBCL) is a lymphoma composed of B cells where the size of malignant cells is equal to or exceeds the size of a macrophage nucleus. DLBCL is the most common lymphoma across the world. All DLBCLs are aggressive lymphomas. The commonest form of DLBCL lacks any special features and is known as DLBCL NOS (not otherwise specified). There four DLBCL subtypes. EBV positive DLBCL of the elderly is a provisional entity in the 2008 WHO classification.

  1. T Cell/histiocyte rich DLBCL (THRLBCL): THRLBCL is a rare variant of DLBCL that is characterised by scattered large B cells that comprise about 10% of the cells in reactive infiltrate that is abundant in T cells with frequent histiocytes.  It resembles Hodgkin’s lymphoma in having a paucity of malignant cells and an abundance of infiltrate. Some TCRLBCL may be arising from progression of nodular lymphocytic predominant Hodgkin’s lymphoma.
  2. Primary CNS DLBCL: Primary CNS DLBCL forms about 90% of primary CNS lymphomas.
  3. Primary cutaneous DLBCL, leg type: Primary cutaneous DLBCL, leg type is a cutaneous lymphoma most commonly arising in the leg. Unlike other DLBCL women are affected more often than men.
  4. EBV positive DLBCL of the elderly

Other forms of DLBCL include those having special anatomical sites (primary mediastinal B cell lymphoma, intravascular lymphoma), histological features (ALK positive large B cell lymphoma, de novo CD5+ large B cell lymphoma) and pathogenesis (large B cell lymphoma arising out of HHV-8 associated Castleman’s disease, pleural effusion lymphoma)

Follicular lymphomas (FL) arise from germinal centres. They have follicle centre (centerocytes/small cell) and large (centroblasts/transformed) arranged at least in a partially follicular pattern. Eighty percent of the patients have the t(14;18)(q32;q21) translocation that results in fusion of immunoglobulin heavy chain gene with BCL2. FL is divided into three categories according to the number of centrblasts. Grade 1-2 FL have 0-15 centroblasts per high power field, Grade 3A FL has >15 centeroblasts per high power field and 3B FL shows solid sheets of centroblasts. Grade 1-2 and Grade 3A FL are indolent lymphomas and Grade 3B is an aggressive lymphoma to be treated as DLBCL.

Small lymphocytic lymphoma (SLL) is a lymphoma that consists small lymphocytes that co-express CD19 and Cd5. It is the nodal counterpart of chronic lymphocytic leukaemia (CLL) and the entity is referred to as CLL/SLL. Patients having lymph node involvement and <5 X 109/L lymphocytes are classified as SLL. Patients with ≥5 X109/L lymphocytes are said to have CLL. The normal counterpart of SLL is the antigen experienced B cell.

Marginal zone lymphomas (MZL) are indolent lymphomas. They are of three types, nodal MZL, extranodal lymphomas of the mucosa associated lymphoid tissue (MALT) and splenic marginal zone lymphomas (SMZL). They arise from post-germinal memory B lymphocytes in the marginal zone of the germinal follicles. About one third of the patients of SMZL do not have somatic hypermutation of the variable regions of the immunoglobulins. The cell of origin is in these SMZL is not known. MZL are peculiar amongst lymphomas in being related to infection. Gastric MALT lymphomas are associated with H. pylori infection, ocular adnexal MALT lymphoma is associated with Chlaymydia psittaci, immunoproliferative small intestinal disease (IPSID) with Campylobacter jejuni, and cutaneous MALT lymphoma with Borrelia burgdorferi. Hepatitis C infection is associated with splenic marginal zone lymphoma.

Mantle cell lymphomas are lymphomas small to medium sized cells that arise form peripheral B cells of the inner mantle zone. It is associated with the t(11;14)(q13;q32) translocation that results in the formation of the IGH@-CCND1 (Cyclin D1) fusion gene. Cyclin D1 can be detected on almost all mantle cell lymphomas by immunohostochemistry.

Burkitts lymphoma (BL) is a lymphoma composed of medium sized cells (nuclei similar to or smaller than histiocytes) that show a diffuse monotonous pattern. The tumour has a very high proliferation index and shows many mitotic figures and a high fraction of apoptosis. It is characterised by translocation that dysregulate the oncogene MYC. These include the t(8;14)(q24;q32) translocation that IGH@ (immunoglobulin heavy chain locus)  to MYC and is the commonest translocation in Burkitt’s lymphoma, the t(2;8)(p12;q24) that translocates the IGK@ (kappa light chain locus) to MYC and t(8;22)(q24;q11) that translocates IGL@ (lambda light chain locus) to MYC. There are two forms of Burkitt’s lymphoma. The Endemic BL occurs in equatorial Africa, affects children and has the EBV genome in majority of the neoplastic cells. The sporadic BL is seen in other parts of the world, is most common in young adults and shows EBV genome only in about 30% of the patients. Sporadic BL is a immunosuppression related malignancy seen in HIV and other forms of immunosuppression.

Lymphoplasmacytic lymphoma is a mature B cell lymphoma that is made of small B lymphocytes and plasmacytoid lymphocytes. These lymphocytes often secret IgM resulting in the syndrome Waldenström macroglobulinaemia. IgM Secretion however in not essential for diagnosis. The normal counterpart of lymphoplasmacytic lymphoma is the post germinal B cell that differentiates into a plasma cell.

Other rarer lymphomas have been described elsewhere (WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues)


Mature T-cell and NK neoplasm

The differentiation of T lymphocytes is not understood as well as that of the B lymphomas.  Clinical picture plays a more important role in the diagnosis of T cell/NK cell lymphomas. T cells carry a more diverse set of function than B lymphocytes. These include cytotoxic functions, aiding other cells of the immune system and regulation of immunity. Many subtypes of T cells are recognised. Like B cells, the T cells have a antigen receptor complex. This consists of and antigen receptor and associated signal proteins. The T cell receptor is made of a pair of chains. There are four T cell receptor chains, α, β, δ and γ. These give rise to two types of T cell receptor the αβ  and δγ. Ninty five percent of the T lymphocytes have the αβ receptors and about 5% of the at T cells have δγ receptors. The δγ T cells and NK cells are a part of the innate immune system. Malignancies of these cells are common children and young adults. These include aggressive NK cell leukaemia, systemic EBV positive lymphoproliferative disease of the childhood, most hepatosplenic T cell lymphomas and δγ-T cell lymphoma.

T cells of the adaptive immune system include naive T cells, helper/regulatory T cells, cytotoxic T cells and memory T cells. Regulatory  cells express CD4. Depending on the cytokine secreting profile these cells are of two types Th1 and Th2. Th1 cells produce IL2 and INFγ that mainly help T cells and macrophages. Th2 cells secrete IL-4, IL-5, IL-6 and IL-10 and mainly help B cell. Follicular helper T cells are T cells that help the germinal centre reaction. In addition to the T cell markers they express germinal centre markers BCL6 and CD10. They also express CD57 and PD-1. Regulatory T cells are cells that suppress immune response. They express CD25.

Lymphomas of the T cells of the adaptive immune system are nodal and occur in adults.

Peripheral T cell lymphoma not otherwise specified (PTCL NOS) is a heterogenous group of malignancies of the peripheral T cells. Its is a basket entity that includes peripheral T cell lymphomas that lack any specific features (unlike the ones listed below). It is the commonest peripheral T cell lymphoma. Gene expression profiling has identified two subtypes of PTCL NOS. Lymphomas arising from the Th1 cells and those arising from Th2 cells.

Anaplastic large cell lymphoma (ALCL) is the second most common T peripheral T cell lymphoma. The normal counterpart of ALCL is not known. ALCL has two subtypes depending on the expression of the anaplastic lymphoma kinase (ALK), ALK+ ALCL and ALK -ve ALCL. These have distinct clinical picture.

Angioimmunoblastic T cell lymphoma (AITL) arises from follicular helper T cells. It usually disseminated at presentation.  It is characterised by generalised lymphadenopathy, systemic symptoms and polyclonal hypergammaglobulinaemia. The patients have immune phenomena including circulating immune complexes, cold agglutinins with haemolytic anaemia, rheumatoid factor and anti-smooth muscle antibodies. These are attributed to polyclonal proliferation of B lymphocytes (which are not the malignant lymphocytes).

Adult T cell Leukaemia/lymphoma is a lymphoma composed of highly pleomorphic lymphoid cells. It is seen in Southwest Japan, Caribbean and parts of Central Africa and is caused by the retrovirus HTLV-I. The clinical types include acute, lymphomatous, chronic and smoldering. Patients often have hypercalcaemia and often have immunodeficiency.

Skin unlike other organs has a higher proportions of T cell lymphomas than B cell lymphomas. These include Mycosis fungoides, Sezary syndrome and the primary cutaneous CD30+ T cell lymphoproliferative disorder, primary cutaneous T cell lymphomas, subcutaneous panniculitis like T cell lymphoma.

Other rare T cell lymphomas include T cell prolymphocytic leukaemia, T-cell Large Granular lymphocytic leukaemia, Extranodal NK/T cell lymphoma, nasal type, enteropathy associated T cell lymphoma and hepasplenic T-Cell lymphoma. A complete list is given elsewhere (WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues)


Hodgkin lymphoma

Hodgkin’s lymphoma is of two types classical and modular lymphocytic predominant. The uncertainty that surrounded the cell of origin of Hodgkin’s lymphoma was ended when microdissected Reed-Sternberg cells were shown to be of B cell origin. The classical Hodgkin’s lymphoma is further divided into lymphocyte rich, nodular sclerosis, mixed cellularity and lymphocyte depletion types.



  1. Elaine S. Jaffe, Nancy Lee Harris, Harald Stein, and Peter G. Isaacson. Classification of lymphoid neoplasms: the microscope as a tool for disease discovery. Blood. 2008 Dec 1; 112(12): 4384–4399.
  2. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues