T-cell prolymphocytic leukaemia associated with immune checkpoint inhibitor (pembrolizumab)
- 1 Internal Medicine Department, Albany Medical Center Hospital, Albany, New York, USA
- 2 New York Oncology Hematology PC, Albany, New York, USA
- Correspondence to Dr Alzira R Avelino; avelina@amc.edu
Abstract
We describe the case of a man in his 60s with squamous cell carcinoma of the lung with brain metastasis treated with pembrolizumab who subsequently developed T-cell prolymphocytic leukaemia. He was transferred to our hospital with worsening dyspnoea, suspected hyperviscosity syndrome and tumour lysis syndrome. He was intubated and admitted to our critical care unit. Emergent leucapheresis was started due to worsening renal function in the setting of tumour lysis and hyperviscosity syndromes. He continued to deteriorate and required continuous renal replacement therapy. Unfortunately, he eventually died from haemodynamic decompensation. There are only a few anecdotal cases pointing at a possible association between the use of immune checkpoint inhibitors and the progression or exacerbation of secondary haematological malignancies. The poor prognosis of these haematological malignancies warrants further investigation to determine if checkpoint inhibitors increase the risk of developing or propagating these potentially fatal diseases.
Background
Pembrolizumab is a member of the class of immunotherapies known as immune checkpoint inhibitors (CPIs) and has been widely used to treat a variety of cancers, including lung cancer. This treatment has been associated with a spectrum of immunotherapy-related adverse events (irAEs), including dermatitis, secondary hypothyroidism, pneumonitis, colitis, as well as infusion reactions. We describe a possible association between the use of pembrolizumab and the progression of secondary haematological malignancy, in this case T-cell prolymphocytic leukaemia (T-PLL). CPIs have been shown to improve survival in their target population, but the suspected potential to induce rapid progression of haematological malignancies can be fatal. This potential association warrants further evaluation to determine whether there is a true relationship between CPIs and development of T-cell lymphoproliferative disorders. If indeed there is an association, then further investigations are warranted to understand the mechanisms behind it to establish a causality hypothesis.
Case presentation
A man in his 60s we are describing here has a medical history of depression, chronic bronchitis, hypertension, atrial flutter, diabetes mellitus type 2, 80 pack-year history of tobacco use, squamous cell carcinoma of the lung with brain metastasis currently on CPI treatment and a recent new diagnosis of T-PLL. He was admitted to an outside hospital two times in the span of 2 weeks with complaints of dyspnoea. First presentation included acute worsening of shortness of breath and narrow complex tachycardia requiring cardioversion by emergency medical services. After discharge he presented a few days later to the emergency department with complaints of worsening dyspnoea and palpitations. He was noted to be in atrial flutter with rapid ventricular response. He was placed on bilevel positive airway pressure and started on rate-control medications. He subsequently was transferred to our hospital for higher level of care and haematological evaluation due to findings of rapidly worsening leucocytosis with lymphocyte predominance.
His home medications included allopurinol 300 mg daily, lurasidone 40 mg daily, vitamin B12 with folic acid (1000–400 µg) lozenges sublingual once daily, prednisone 50 mg two times per day (started on recent previous admission for chronic obstructive pulmonary disease exacerbation), Colace 100 mg two times per day, albuterol 90 µg HFA 2 puffs by inhalation every 4 hours as needed, sotalol 80 mg two times per day, metformin 1000 mg tablet daily.
His oncological history is notable for a stage IIIB left upper lobe squamous cell carcinoma of the lung. He received two cycles of neoadjuvant chemotherapy with cisplatin and etoposide concurrent with radiation followed by left extrapleural pneumonectomy including partial pericardiectomy. A year later he developed recurrence of his lung cancer with a left medial frontal brain metastasis for which he received stereotactic radiosurgery. Molecular phenotyping of his resected malignancy demonstrated programmed death-ligand 1 (PD-L1) expression of greater than 90%. Therefore, he was started on palliative intent treatment with pembrolizumab as supported by the KEYNOTE-024 trial.1 He had received maintenance therapy with pembrolizumab for 19 months and the last dose was administered 2 weeks prior to presentation to our hospital.
A little over a year after starting pembrolizumab the patient developed persistent leucocytosis and hepatosplenomegaly. He underwent a bone marrow biopsy, which was consistent with a T-cell prolymphocytic polymorphic leukaemia diagnosis. The patient was scheduled to stop pembrolizumab and start alemtuzumab for treatment of this newly diagnosed leukaemia prior to this admission but the treatment had not yet been initiated.
Investigations
At the time of initial diagnosis with T-PLL the patient was noted to have persistent leucocytosis of about 20.4×103/µL (4–9×103/µL) on complete blood count (CBC) with an absolute neutrophil count of 12.0×103/µL (1.5–5.2×103/µL) and absolute lymphocyte count of 7.3×103/µL (1.1–3.9×103/µL). CBC also revealed a haemoglobin of 12.8 g/dL (13.6–16.7 g/dL), haematocrit 40.50% (40.0%–49.0%), platelet count of 315×103/µL (130–350×103/µL). The diagnosis of T-PLL was achieved by the assertion of two major criteria, known as T-PLL defining characteristics (T-cell clonality by flow cytometry and T-PLL phenotypical cells on blood and bone marrow) and one minor criterion (abnormality of chromosome 11q involving the ATM gene).2 Flow cytometry in October of 2020 revealed increased CD4/CD8 ratio with expansion of the CD4 positive T cells. Molecular genetics testing for T-cell receptor gamma gene rearrangement was suggestive of monoclonality. Bone marrow biopsy showed atypical CD4 positive T cells comprising 20% of the marrow cellularity with increase in the prolymphocytic forms consistent with T-cell lymphoproliferative disorder (figure 1). Cytogenetics analysis and fluorescence in situ hybridisation (FISH) of the peripheral blood revealed a complex, unstable karyotype with multiple structural and numerical abnormalities that included deletion of 8p and 11q involving the ATM gene. There was no rearrangement of the TCL1 gene on FISH analysis or presence of T-cell receptor rearrangement.
H&E (A) and CD3 stain (B) of bone marrow biopsy demonstrate CD3 positive T cells comprising 20% of the marrow cellularity (400×, A) with circulating prolymphocytic forms (1000×, B). The T cells express CD4, CD2, CD3, CD5, CD7, CD52, CD25(dim) and are negative for CD8, CD30 and CD279a.
On presentation to our hospital initial CBC showed a white cell count of 475.5×103/µL (4–9×103/µL) with an absolute neutrophil count of 71.3×103/µL (1.5–5.2×103/µL) and absolute lymphocyte count of 9.5×103/µL (1.1–3.9×103/µL) with 361.4×103/µL atypical immature T-PLL cells. Haemoglobin of 10 g/dL (13.6–16.7 g/dL), haematocrit 32.4% (40.0%–49.0%), platelet count of 115×103/µL (130–350×103/µL). His metabolic panel showed hyperkalaemia with potassium of 7.1 mEq/L (3.4–5.2 mEq/L), lactate dehydrogenase of 3211 IU/L (90–255 IU/L), phosphorus 4 mg/dL (2.4–4.7 mg/dL) and lactic acid 2.97 mmol/L (0.5–2.2 mmol/L).
CT angiogram was negative for pulmonary embolism. He was tested for SARS-CoV-2, which was negative, influenza and respiratory syncytial virus were also negative. Blood cultures were subsequently negative.
Differential diagnosis
Differential diagnoses were contemplated but due to rapid rise in white blood cell count and a known underlying haematological malignancy there were not many alternative hypotheses other than rapid progression of T-PLL. Additional differential diagnoses would include a possible infection, which can worsen leucocytosis in many haematological malignancies. However, complete infectious workup returned negative. Moreover, the degree of leucocytosis would make this differential very unlikely. Progression to active T-PLL was our main diagnosis, supported by the acute presentation, degree of haematological abnormality and lack of other conceivable differential diagnoses.
Outcome and follow-up
On transfer to our hospital, he was noted to be in respiratory failure requiring intubation. He was admitted to the critical care unit with haemodynamic instability requiring intravenous vasopressor support. He had severe leucocytosis causing tumour lysis and hyperviscosity syndromes. He was started on leucapheresis and continuous renal replacement therapy for worsening acute renal failure and hyperkalaemia. We started the process to emergently acquire alemtuzumab to administer in the hospital to treat active T-PLL. Unfortunately, the patient continued to deteriorate with no promise of improvement. Ultimately, the patient was made comfort care measures only by healthcare proxy and passed away shortly after in the presence of his family.
Discussion
Programmed cell death protein 1 (PD1) is a protein expressed on all T cells during activation, and it functions as a ‘brake’ on the effector T-cell response. It is usually highly expressed in acute or chronic infections and in malignancies.3 T cells that express PD1 are inactivated by engaging with tumour cells expressing programmed cell death ligand 1/2 (PD-L1/2), thus hindering their cytotoxic effect.3 Cytotoxic T-cells function and proliferation can be recovered by the addition of an immune CPI to disrupt binding of PD1/PD-L1/2, and hence, preventing an ‘exhausted’ T-cell phenotype in the tumour microenvironment.1 3
Pembrolizumab is an anti-PD1 monoclonal antibody which inhibits PD1 activity by blocking the binding of PD1 proteins on T cells with PD-L1 expressed on tumour cells and other immune cells within the tumour microenvironment. It has been used to treat a variety of non-small cell lung malignancies, including both non-squamous and squamous cell lung cancers.1 4 5 This agent can be used alone to treat advanced squamous cell lung cancers overexpressing PD-L1 at 50% or higher.1 It can also be used in combination with cytotoxic platinum-doublet chemotherapy with paclitaxel or nab-paclitaxel.5 There have been numerous reported immunotherapy related adverse events (irAEs) associated with this drug class, including dermatitis, secondary hypothyroidism related to hypophysitis or following hyperthyroidism, pneumonitis, colitis and infusion reactions.5 6
We report a possible association between pembrolizumab therapy and the progression of T-PLL. T-PLL is a rare post-thymic haematological malignancy that usually affects older individuals with an average age of 65 years.7 The diagnosis requires the presence of three major criteria (>5×109/L cells with T-PLL phenotype in peripheral blood or bone marrow; T-cell clonality by PCR for T-cell receptor genes (TRB/TRG) or by flow cytometry and abnormalities of 14q32 or Xq28 or overexpression of T-cell leukemia/lymphoma1 (TCL1) family genes; TCL1A, TCL1B (TCL1/MTCP1-like 1 (TML1)) or mature T-cell proliferation (MTCP1) or the first two major criteria and one minor criterion (abnormalities involving the chromosome 11 (11q22.3; ATM), abnormalities involving chromosome 8, 5, 12, 13, 22 or complex karyotype; involvement of T-PLL specific site).2 If the third major criterion is not present in a case then it is classified as TCL1-family negative T-PLL.2 Clinically, most patients will present with generalised lymphadenopathy and hepatosplenomegaly.8 Typical haematological findings include marked and rapidly rising leucocytosis, thrombocytopenia and anaemia.9 A diagnostic feature of T-PLL is the presence of prolymphocytes in peripheral blood and bone marrow.7 10 T-PLL is a very aggressive haematological malignancy with poor prognosis.11 Treatment for T-PLL is reserved for patients who demonstrate disease progression to ‘active’ disease. The guidelines from the T-PLL International Study Group (TPLL-ISG) recognise the following criteria as markers of active disease: disease-related constitutional symptoms, symptomatic bone marrow failure, rapidly enlarging lymph nodes, spleen, and liver, increasing lymphocytosis, or extra nodal involvement.2 Initial treatment includes the use of anti-CD52 monoclonal antibodies like alemtuzumab, and in some cases purine analogues.12 13 Moreover, allogeneic stem cell transplant in the patients who achieved remission results in longer survival.14
Authors acknowledge that T-PLL can have an underlying latent phase but in our case there was no haematological abnormality of concern before initiating CPI. Hence, T-cell subset analysis and clonality determination check was not deemed necessary. Subsequently as the haematological abnormality was identified further investigation led to official diagnosis of T-PLL based on diagnostic criteria described above.
There have not been many reported associations between the use of CPIs and any development, progression or exacerbation of secondary haematological malignancies. Anand et al have described the development of a secondary peripheral T-cell lymphoma not otherwise specified in an elderly man being treated with pembrolizumab for undifferentiated carcinoma of epithelial origin.15 They also provide a review of the Food and Drug Administration adverse events which showed 12 reported T-cell–related neoplasms with the use of CPIs, of which three were associated with pembrolizumab.15 Angioimmunoblastic T-cell lymphoma has also been reported in a patient who received ipilimumab followed by maintenance pembrolizumab for metastatic spindle cell melanoma.16
Although the development of secondary T-cell malignancies has not been well described in patients being treated with CPIs, this report and others described above support considering further investigations in this relationship. Preclinical animal studies have revealed that unrestricted T-cell activation following an oncological insult can lead to T-cell proliferation and potentially T-cell malignancies. For example, Wartewig et al have described a tumour suppressing role in PD1 proteins in T cells. They used a mouse model with a T-cell lymphoma carrying a specific oncogenic mutation. In models with biallelic PD1 deletion there was unregulated proliferation of T cells with the specific oncogenic mutation, but in the presence of PD1 the proliferation was diminished. In conclusion, PD1 deletion leads to unrestricted T-cell proliferation after an oncogenic insult and results in aggressive lymphoma.17 Our case here raises a concern whether a similar hypothesis can be postulated as a potential causality.
To our knowledge, there has not been any reported long-term data indicating that CPIs can induce unregulated mutation and rapid proliferation of clonal cells. However, this report illustrates a possible association between rapid progression of T-PLL in the presence of PD1 inhibition. This association, if confirmed, will require further investigation including the identification of mechanisms behind how leukemogenesis can occur following disruption of the interaction of PD1 and PD-L1 in the tumour microenvironment.
Learning points
-
As immune checkpoint inhibitor therapy is more widely used to treat cancer patients, there remains the need for continued monitoring for long-term toxicities as well as identifying rare but potentially fatal clinical outcomes as discussed in our case report.
-
There may be an association between the use of CPIs and the development of secondary T-cell malignancies; however, further studies are warranted to confirm a ‘cause and effect’ and if this is the case, basic science mechanisms behind these processes need to be investigated.
-
There remains several unanswered questions about the long-term sequelae of PD1 and PD-L1/2 inhibition on neoplastic and non-neoplastic cells.
Ethics statements
Patient consent for publication
Acknowledgments
Bone marrow pathology slides were obtained with the help of Dr Jacqueline Cortazar from the pathology department at St Peter’s Hospital in Albany, NY.
Footnotes
-
Contributors AA completed the initial draft of the case including literature review, as well as subsequent editing and drafts. OE did additional literature review as well as document revisions/editing for subsequent drafts. Additional review, editing and mentorship provided by SW and MR.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2022. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer