Durable responses with ALK inhibitors for primary refractory anaplastic lymphoma Kinase-positive large B-cell lymphoma

Anaplastic lymphoma kinase (ALK)-positive large B-cell lymphoma (LBCL) was first identified in 1997¹ and was defined as a unique subtype of diffuse large B-cell lymphoma (DLBCL) by the World Health Organization in 2008.² Unlike ALK-positive anaplastic large-cell lymphoma (ALCL), ALK-positive LBCL has both immunoblastic and plasmablastic cells with a unique immunohistochemical pattern that is negative for B cell (CD20, CD79a), T cell (CD2, CD3), and CD30 markers, but has expression of epithelial membrane antigen (EMA), CD38 and CD138. Another feature common to ALK-positive LBCL is diffuse and/or sinusoidal invasion.¹ Prior case reports have detected t(2;17)(p23;q23) involving clathrin heavy-chain gene (CLTC) at chromosome 17q23 and ALK gene at chromosome 2p23, thereby creating CLTC-ALK rearrangement^3,4 or presence of t(2;5)(p23;q35) creating NPM-ALK fusion proteins.⁵ 

ALK-positive LBCL accounts for <1% of DLBCL and has a male predominance with median age at diagnosis ∼35 to 38 years.^2,5 In addition to being a rare form of non-Hodgkin lymphoma, ALK-positive LBCL has poor outcomes with median overall survival (OS) ∼1 year among those with advanced-stage disease^6,7 and 5-year OS of 8% in those with stage III/IV disease.⁵ Survival is poor as these patients have minimal response to conventional systemic chemotherapies, such as Cytoxan, H-adriamcyin, O-vincristine, Prednisone (CHOP), Cytoxan, H-adrimamycin, O-vincristine, Etoposide, Prednisone (CHOEP), Etoposide, Prednisone, O-vincristine, Cytoxan, H-adriamycin (EPOCH), and Hyper-Cytoxan, Vincristine, Adriamycin, Dexamethasone (HyperCVAD).⁵ 

Crizotinib is an oral, small-molecule tyrosine kinase inhibitor targeting ALK, MET, and ROS1 with preclinical activity in ALK+ anaplastic large cell lymphoma (ALCL).⁸ Although it has approval for ALK rearranged advanced non-small cell lung cancer, it is actively being explored in other malignancies with ALK positivity. In a phase 1b trial,⁹ crizotinib was administered to patients with diagnosis of advanced malignancies with ALK translocation, inversion, mutation, or amplification, such as ALCL, inflammatory myofibroblastic tumors, and non-Hodgkin lymphoma. One of the patients in this trial with ALK-positive LBCL had stable disease for approximately 4 years on crizotinib. Although crizotinib is an option for relapsed/refractory (R/R) ALCL,^10,11 it is not currently approved for ALK+ LBCL.

The exact pathogenesis that drives ALK+ LBCL is unclear. Preclinical data with mice models have suggested that overexpression of NPM-ALK or TPM3-ALK oncogenes may halt early B-cell differentiation and lead to ALK+ B-cell lymphoproliferative disorders, which could be reversed with use of ALK inhibitors.¹² To date, there have only been a few case reports involving the use of crizotinib for R/R ALK+ LBCL with responses lasting <1 month.^13-15 There has only been 1 case report using alectinib for ALK+ LBCL but was for a pediatric patient and was used in combination with chemotherapy.¹⁶ Recently, a study found promising efficacy of higher-potency ALK inhibitors (alectinib and lorlatinib) when studied in xenograft models derived from patients.¹⁷ Based on this preliminary data, 4 patients with R/R ALK+ LBCL (75% had progressed on crizotinib) were treated with alectinib at 600mg twice daily with 3 complete responses (CR) and 1 partial response (PR). Two of these patients eventually underwent myeloablative allogenic stem cell transplant, whereas the other 2 patients had progressive disease (PD). Of the patients with PD, 1 was treated with lorlatinib 100mg daily and attained CR.

Here, we report 2 cases of patients with R/R ALK+ LBCL who were treated with ALK inhibitors.

A male aged 23 years with asthma noticed left neck swelling which progressively worsened with associated fevers, chills, night sweats and weight loss. Computed tomography (CT) imaging of neck, chest, abdomen, and pelvis revealed disease localized to bilateral cervical lymph nodes (LN). Excisional biopsy of left cervical LN was notable for ALK+ LBCL with immunohistochemistry (IHC) stains showing positive ALK (cytoplasmic and golgi pattern), EMA, MUM1, weak CD45, weak CD79a, and CD138 (subset), with remaining stains shown in Table 1. Positron emission tomography-CT (PET/CT) revealed disease in palatine tonsil, adenoids, base of tongue, and left cervical and supraclavicular (SCV) LN. Bone marrow biopsy was negative for lymphoma involvement. He completed 3 cycles of CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone). Subsequent PET/CT revealed Deauville score of 4 in left cervical and SCV LN, thus the patient was started on crizotinib 250mg twice daily about 3 weeks after CHOP. He subsequently received localized radiation therapy to left cervical LN and a follow-up PET/CT after 3 months showed near CR. Surveillance CT scans every 6 months have shown ongoing CR while on crizotinib therapy. He has maintained a durable response to crizotinib for at least 6 years. His course has only otherwise been complicated by an incidental finding of a left subsegmental pulmonary embolism (PE), which was found on a surveillance scan in December 2019 and was managed with 3 months of rivaroxaban.

A male aged 46 years with hypertension and obstructive sleep apnea presented with 2-months of an enlarging left neck mass. He had no constitutional symptoms at the time. CT neck and chest were notable for enlarged left cervical and supraclavicular (SCV) LN with maximum diameter of 10.7 cm. Core biopsy of left cervical LN revealed ALK+ LBCL with IHC positive for ALK (granular cytoplasmic pattern), CD45, CD138, weak PAX5, and weak CD30; and IHC negative for CD20, CD3, CD5, CD8, S100, and MCK. PET/CT revealed an FDG-avid conglomerate left neck mass extending into the upper mediastinum and SCV region (SUV max 29) along with a left parapharyngeal lesion. He was, therefore, diagnosed with ALK+ LBCL, stage II, bulky disease. The patient received 1 cycle of CHOP. Within 15 days, he noticed worsening left neck swelling that was confirmed as clinical progression. Despite a dexamethasone burst (40mg daily for 4 days), he had no symptomatic relief and was urgently treated with radiation to the left neck mass (50 Gy/ 20 fractions). Within a few days after the completion of radiation therapy, he noticed a new left axillary LN. CT imaging confirmed enlarged bilateral axillary lymphadenopathy with reduction in left SCV LN. For systemic relapse, we were able to get approval for crizotinib 250mg twice daily, which he began approximately 1 month after radiation completion. The patient noted a drastic improvement in axillary lymphadenopathy within days. Upon initiation of crizotinib, there was a transient rise in liver function tests (AST 151 U/L, ALT 323 U/L) that resolved with hydration and conservative management. PET/CT completed approximately 8 weeks after radiation therapy noted a PR. However, about 2 months after crizotinib use, he reported enlarging bilateral axillary lymphadenopathy. PET/CT at this time confirmed interval PD in bilateral axilla along with a new left pleural effusion and minimally FDG-avid pulmonary nodules. He was referred back to radiation oncology and began localized radiation to the left axilla (50 Gy/ 20 fractions) and was switched to alectinib 600mg twice daily around the same time. Post-treatment PET/CT has revealed resolution of left axillary LN, ongoing reduction in left cervical LN, and mixed response within right axillary LN. He maintained PR on alectinib for 6 months but subsequently developed relapsed disease predominantly in the right axilla and right SCV nodes. Right axillary biopsy confirmed ALK+ LBCL with immunohistochemical stains negative for CD30 and CD19. TEMPUS testing from lymph node biopsy was only notable for CLTC-ALK rearrangement and copy number loss of ATM. He began focal radiation to right axilla, and for systemic therapy, was switched to lorlatinib 100 mg daily which he has been on for at least 6 months with evidence of complete metabolic response on PET. His course was otherwise complicated by coronavirus disease 2019 (COVID-19) with incidental findings of a left lower lobe subsegmental PE that was managed with apixaban.

In summary, we report the first case series of adult patients with ALK+ LBCL who were treated with ALK inhibitors and attained meaningful and durable clinical responses. Both our patients had diseases that were resistant to standard-of-care treatment with CHOP. Initiation of an ALK inhibitor led to a rapid clinical response in both patients as demonstrated in Figure 1. Our first patient from case 1 has done exceptionally well with crizotinib as demonstrated by ongoing CR for >6 years without any dose-limiting toxicities. In contrast, our second patient, after an initial response, had PD after 3 months of crizotinib therapy and, thus, was switched to a second-generation ALK inhibitor, alectinib. He tolerated this well with response lasting 6 months and now in a CR while being treated with lorlatinib currently again in a CR after progressing while on alectinib.

Preclinical in-vitro studies of ALK+ LBCL have demonstrated that ALK inhibitors can delay tumor growth and lead to tumor regression.^18,19 Experimental data has also suggested that different ALK fusion genes may have differential sensitivity to ALK inhibition.^18,20 This could potentially explain why one of our patients responded well to crizotinib, whereas the other patient progressed on this drug and has required treatment with newer generation ALK inhibitors.

Based on our promising findings, further studies should be performed to investigate the efficacy and tolerability of ALK inhibitors as a potential treatment option for patients with ALK+ LBCL, as it could drastically impact the morbidity and mortality of this disease.

Acknowledgment: This work was financially supported by the Ruth L. Kirschtein National Research Service Award by the National Institutes of Health: 5T32CA009357-39.

Contribution: R.T. wrote and prepared the manuscript, and T.P. helped write and edit the manuscript.

Conflict-of-interest disclosure: No conflicts of interest to disclose.

Correspondence: Tycel J. Phillips, Division of Hematology, Department of Hematology, City of Hope, 1500 East Duarte Road, Duarte, CA 91010; e-mail: tphillips@coh.org.