HbA1c is a glycosylated hemoglobin (Hb) adduct that is well-accepted as a versatile marker of hyperglycemia in diabetes mellitus (DM). It can be used to screen, diagnose, and monitor DM patients so that optimum treatment can be given to slow down disease progression. However, it may not represent the actual glycemic control status in certain patients. The presence of Hb variants, diseases that impair red blood cell (RBC) survival (such as hemolytic anemia), as well as the administration of certain drugs that may affect RBC integrity (e.g., dapsone, hydroxyurea) or vitamins (e.g., vitamins C and E) are known factors that may adversely affect HbA1c levels. Here, we report a rare case of undetectable HbA1c associated with hyperleukocytosis secondary to T-cell prolymphocytic leukemia (TPLL).

A 54-year-old Malay gentleman with underlying hyperlipidemia and type 2 diabetes mellitus (T2DM) on insulin therapy since 2004 initially presented with a few weeks’ history of lethargy and epigastric discomfort associated with abdominal distention and pedal edema. He was admitted for further assessment. The initial workup, including a full blood count (FBC), showed that the total white cell (TWC) count was markedly raised (574.60 × 10⁹/L), with lymphocytes predominating (558.42 × 10⁹/L [97.2%]), a red blood cell (RBC) count of 3.81 × 10⁹/L, Hb = 11.3 g/dL, and platelets = 81 × 10⁹/L. A full blood picture (FBP) revealed significant hyperleukocytosis, with 98% abnormal lymphoid cells (Figure 1). Based on these features, lymphoproliferative disorder needed to be considered. To explain the patients’ white cell count of over 50,000 × 10⁹/L, a bone marrow aspiration and trephine (BMAT) biopsy and flow cytometry immunophenotyping were performed, revealing TPLL. The patient was started on induction chemotherapy with 3-cycle fludarabine, mitoxantrone, and cyclophosphamide, as well as prophylactic allopurinol, hydroxyurea, and hyperhydration as a preventive strategy against tumor lysis syndrome (TLS). In view of the patient’s comorbidity of T2DM, fasting blood glucose (FBS) and HbA1c were measured to assess glycemic control. The FBS value was 11.2 mmol/L. However, surprisingly, no HbA1c peak was detected in the initial run using the capillary electrophoresis (CE) method (Figure 2).

Thus, the sample was treated by removing the plasma, and the remaining RBCs were washed 3 times with saline (the sample was centrifuged after each wash step). The analysis was repeated, and the result is shown in Figure 3.

Figure 1

Figure 2

Figure 3

DM is a common occurrence in cancer patients due to the effects of cancer treatment regimes. Many patients with pre-existing DM undergoing cancer treatment develop iatrogenic hyperglycemia most commonly associated with steroid administration1. This may pose a significant challenge in monitoring the patient’s actual glycemic control status. Clinical management should be tailored to the individual patient to maintain euglycemia and prevent further complications2. In the present case, the patient was diagnosed with TPLL on top of having T2DM and being on insulin therapy for the past 20 years; this presentation posed a challenge to the managing team.

TPLL is characterized by the malignant proliferation of mature T lymphocytes. The morphology is described as hyperleukocytosis with the marked proliferation of small- to medium-sized prolymphocytes, showing distinct cytoplasmic blebbing, similar to our case. The cells in B- and T-CLLs are small and mostly well-differentiated. Immunophenotypically, they are positive for CD7, CD2, CD5, and CD3, with CD52 usually also exhibiting bright expression. Patients will usually be given high-dose chemotherapy, and in certain cases, hematopoietic stem cell transplantation may offer a curative outcome. Nevertheless, it is generally resistant to conventional chemotherapy and has a poor prognosis3.

The presence of hyperviscosity and leukostasis syndromes in TPLL is uncommon. To increase whole-blood viscosity, a cell count of at least 1,000,000 cells/µL is needed to reach a leucocrit of 20%4. Hyperleukocytosis occurs when the white blood cell (WBC) count exceeds 100,000 cells/µL. When hyperleukocytosis becomes clinically evident, it is known as leukostasis. It can result in complications such as TLS and disseminated intravascular coagulopathy (DIC)5.

Various methods can be used to improve patients’ outcomes in cases of symptomatic hyperleukocytosis. These include induction chemotherapy, supportive treatments, and cytoreductive therapy using corticosteroids, leukapheresis, and chemotherapeutic agents such as hydroxyurea6.

At presentation, the current patient had hyperleukocytosis without leukostasis. He was given fluid therapy, allopurinol, and hydroxyurea as preventive measures against TLS at the commencement of induction chemotherapy. In addition to the assessment of liver and renal function and electrolyte abnormalities, HbA1c was measured as part of the monitoring process in view of the patient’s comorbidity. However, the initial HbA1c result was unreportable as no HbA1c peak was detected by the assay. HbA1c was measured using CE, which is reported to have better resolution and less interference from interfering substances. In the present case, hydroxyurea was given to the patient; this drug is known to interfere with HbA1c measurement. The drug causes a shift from HbA to HbF, leading to a lower HbA1c level7, 8. The drug should thus be withheld to stop the alteration process to measure the HbA1c level9. However, the false HbA1c result of the initial run in this report is not likely to be due to interference with hydroxyurea as the HbA1c peak was detected after washing the sample with saline. If hydroxyurea had affected the HbA1c peak, the result should have remained the same even after the washing step.

Other causes of an absent HbA1c peak include the presence of Hb variants that can interfere with HbA1c measurement methods. For example, HbA synthesis is reduced in β-thalassemia and hereditary persistence of fetal Hb (HPFH), causing falsely low HbA1c regardless of the analytical methods used10. Drugs that can cause hemolysis, like dapsone and certain antiretroviral agents, can also result in spuriously low HbA1c levels11.

The absence of an HbA1c fraction in this patient is postulated to be due to accelerated migration caused by the presence of some unknown molecules secreted by the malignant cells. It is unlikely to be due to factors endogenous to the erythrocytes since the HbA1c peak was recognizable after the saline wash12.

Hyperleukocytosis is a potential cause of interference in HbA1c measurement, possibly due to proteins or other substances secreted by the malignant cells. As presented, in our patient with hyperleukocytosis, no HbA1c peak was detected in the pre-treatment sample. This hypothesis arises based on the observation that washing the plasma out of the sample could eliminate the interference. Unfortunately, the exact mechanism that underlies the condition is still unclear, and further studies are needed to identify its true cause.

BMAT: bone marrow aspiration and trephine biopsy CE: Capillary Electrophoresis DIC: Disseminated Intravascular Coagulopathy DM: Diabetes Mellitus FBP: Full Blood Picture FBS: Fasting Blood Glucose HbA1c: Glycosylated Haemoglobin Hb: Haemoglobin HPF: High Power Field HPFH: Hereditary Persistence of fetal Hb RBC: Red Blood Cell T2DM: Type 2 Diabetes Mellitus T-PLL: T-cell prolymphocytic Leukemia TWC: Total White Cell Count TLS: Tumor Lysis Syndrome

We thank Dr. Aniza Mohammed Jelani, Head of the Department of Chemical Pathology, for her support.

Abdul Razak AA, Tuan Ismail TS, Zulkeflee RH, Ab Rahim SN, Zulkeflee HA, and Wan Nik WNFH, contributed equally to this work; Wan Nik WNFH supervised and directed the focus of this review; Abdul Razak AA, Tuan Ismail TS, Zulkeflee RH, Ab Rahim SN, Zulkeflee HA, Wan Nik WNFH performed the literature review. All authors have reviewed and agreed upon the final version of this case report.

None.

Data and materials used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Not applicable.

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

The authors declare that they have no competing interests.