• 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • At this time progressive hilar disease was


    At this time, progressive hilar disease was biopsied, confirming TTF-1 positive adenocarcinoma, ALK positive by IHC (ALK1); next generation sequencing (NGS) further confirmed an EML4-ALK fusion with no additional kinase mutations, but sample was insufficient to identify other somatic variants. He then commenced brigatinib with initial transient symptomatic improvement but subsequent further bone disease progression confirmed on MRI spinal imaging following three cycles of brigatinib. Due to clinical deterioration, unsuitable tissue to rebiopsy and unavailability of ctDNA NGS, he next switched to lorlatinib with a marked symptomatic improvement during the first cycle of treatment. Lorlatinib was well tolerated with no required dose-reductions. Response evaluation CT-imaging following two months of lorlatinib demonstrated a partial response (Fig. 1(e) and (f)). Response continued for an additional five months with CT surveillance then demonstrating progression due to new liver metastases. Image-guided liver-biopsy identified TTF-1 negative NSCLC, expressing CD56, synaptophysin, and chromogranin, interpreted as NSCLC with neuroendocrine differentiation (Fig. 2). PDL1 testing was negative (22C3) and ALK immunohistochemistry (D5F3) was positive, validated by FISH (Vysis break apart). NGS confirmed an EML4-ALK fusion, but also an L1196 M gatekeeper mutation, and CDKN2A homozygous deletion, also confirmed by FISH, with no RB1 loss. The patient commenced carboplatin-vinorelbine chemotherapy for two cycles but clinically deteriorated through this and rapidly thereafter, consistent with clinical progression. The patient’s care was transferred to the Acarbose palliative care services, and he subsequently passed away. The sequence of anti-cancer treatments is summarised in Table 1.
    Discussion We report a case of neuroendocrine carcinoma transformation as a mechanism of resistance to the next generation ALK-inhibitor lorlatinib, following three previous three lines of ALK-targeted therapy. To the best of our knowledge, this is the first report of neuroendocrine carcinoma transformation as a resistance mechanism to lorlatinib following three generations of ALK inhibition. SCLC transformation has been reported following progression on crizotinib alone [6,11] post alectinib after crizotinib [[12], [13], [14]], post ceritinib after crizotinib [7], and following lorlatinib after crizotinib [8]. Consistent with all these cases where re-biopsied SCLC tissue was examined [6,[11], [12], [13], [14]], ALK rearrangement was retained in the transformed SCLC – indicating true transformation, and not outgrowth of an existing SCLC clone. Our case therefore further confirms neuroendocrine carcinoma transformation as a resistance mechanism that is not exclusive to crizotinib use. (ceritinib and alectinib) ALK inhibitors [3]. Further mechanisms of resistance include ALK gene amplification [4,15], activation of different kinases such as EGFR [15], KIT [15], nonreceptor tyrosine kinase (SRC) [16], and insulin like growth factor 1 receptor [17]. These resistance mechanisms are thought to occur due to maintained activation of downstream ERK and/or PI3K–AKT signaling despite ALK inhibition. Ou et al [8] recently reported a complimentary similar case of the ALK G1202R solvent-front mutation and SCLC transformation as resistance mechanisms to second generation ALK inhibitors, without prior exposure to crizotinib. We report the first case of dual occurrence of the ALK L1196 M ALK gatekeeper mutation and neuroendocrine transformation as a resistance mechanism to lorlatinib. Our case, therefore, again highlights the risk of relying on liquid biopsies in the clinic for oncogene-addicted NSCLC, as this would have simply detected the L1196 M ALK variant. Whilst L1196 M has been identified as an acquired variant associated with resistance to crizotinib, ceritinib, and alectinib, but not brigatinib, the identification of this variant after brief exposure to brigatinib and longer exposure to lorlatinib suggests the L1196 M variant was associated with resistance to lorlatinib. This is, however, unusual, since in preclinical models this mutation results in lorlatinib sensitivity [10]. Nevertheless, we cannot exclude a competing role in acquired resistance of the CDKN2A loss identified, or that L1196 M developed on brigatinib.