PNH is a nonmalignant clonal disorder that originates from a singular stem cell that acquires a mutation in a gene called PIG-A. As such, PNH is not inherited but rather is acquired at some point in a person’s life. The PIG-A mutation leads to a partial or absolute deficiency of GPI-anchored proteins (GPI-APs) in the stem cell and in its progeny. The resulting cell line will bear the same genetic and phenotypic defects and can coexist with normal unmutated clones.

Flow cytometry is the preferred technique for the diagnosis of PNH and is performed to evaluate for the presence of GPI-linked antigens on blood cells. This testing is used to classify PNH, inform its treatment, monitor treatment response, and detect disease progression. Using flow cytometry, PNH red blood cells are classified as type 1, 2, or 3. Type 1 cells have a normal expression of GPI-AP, type 2 cells have a partial deficiency of GPI-AP, and type 3 cells lack all GPI-APs. There are some nuances to appreciate with these categories. For example, a patient with PNH may have a small fraction of cells that are genotypically PNH cells but have normal GPI-AP levels arising through passive transfer from unaffected cells.

An increased propensity for thrombotic events in PNH is well known, and clone size may be one of several key aspects of thrombotic risk in PNH. At the ASH conference, Gurnari et al presented data from their analysis of clinicomolecular risk factors for thrombosis in a large real-life cohort of patients with PNH (abstract 3818). Interestingly, researchers reported a trend for a higher incidence of thrombosis among patients with type 2 PNH as compared with those with type 3 disease. One might theorize that individuals with type 3 PNH, completely lacking GPI-APs, may be at a higher risk for developing PNH-related complications, but there are intricacies that make for further clinical heterogeneity. For example, patients presenting with type 3 PNH often have a very small number of aberrant cells in circulation because these cells are lysed almost as soon as they are produced; in contrast, clone sizes may be allowed to grow quite large in type 2 PNH due to incomplete clonal lysis. Thus, a patient presenting with type 2 PNH in a hemolytic event might have a greater risk of thrombosis than a patient presenting with type 3 PNH, because there are more PNH cells to lyse and it is the lysis of these larger clones that is associated with thrombosis. 

The same concern may apply to breakthrough hemolytic events among patients who are treated with anticomplement therapies (eg, during complement-amplifying conditions such as infection). Additionally, now that we have very good complement inhibitors, treatment is associated with an increase in the fraction of PNH red blood cells because they are protected from lysis. If, for some reason, the patient stops treatment suddenly or if they have an inability to receive the drug, the potential for hemolysis could become very dramatic because there are many susceptible cells. The bottom line, in my view, is that, if we want to protect PNH cells, having large amounts of susceptible cells is actually a measure of how effective or successful we are in blocking complement.

Another abstract by Gurnari and colleagues from ASH 2022 touched on the clonal evolution of hematopoietic stem cells in aplastic anemia (AA) and PNH (abstract 2564). We should recognize that clonal evolution is variable in PNH and that there is tremendous diversity in the clinical behavior of the disease. In AA and PNH, the acquisition of somatic PIG-A mutations is thought to provide a selection advantage to the mutated stem cells whereby they evade immune attack.

There are different scenarios to consider, one of which is that a neoplastic process may involve a linear acquisition of additional mutations within a PNH clone. In abstract 2564, researchers were interested in myelodysplastic syndromes arising from AA and PNH and the genomic landscape that is typical of AA/PNH clonal evolution. Within a PNH clone, along with the founder PIG-A mutation, you can acquire additional mutations; the more mutations you acquire, the more that clone will resemble myelodysplastic syndromes. One can envision that some mutations may create a disconnect with the immune milieu, contributing to immune evasion and a competitive advantage. Some patients progress quickly while others progress more slowly. This raises the question of which features the mutant clone might be acquiring that make it less susceptible to immune attack. As noted by Kawashima et al, the expression of immunomodulatory molecules (eg, checkpoints) differs between PNH cells and phenotypically normal cells, with these differences possibly explaining the diversity of the clinical expression of the disease (abstract 2565). 

Another possibility is that a person has damaged bone marrow at multiple sites and that the clone that becomes neoplastic is not within the PIG-A mutant clone; it is more like a side-clone. In that case, from the start, the neoplastic clone does not rely on circumventing extrinsic immune pressure. And the proliferative drive is more related to intrinsic properties of the clone that are conveyed by the additional mutations. In such situations, a side-clone may outcompete PNH, and sometimes this manifests itself as a PNH “cure” (ie, PNH cells start disappearing because another clone outcompetes the PNH clone).