Patients with metastatic lung adenocarcinoma (mADC) harboring EGFR-activating mutations can benefit from first-line Osimertinib, but acquired resistance inevitably occurs. Different resistance mechanisms, on- and off-target, have been described. Here, we evaluated the prevalence of phenotypic transformation as a resistance mechanism in a consecutive series of EGFR-mutated mADC, diagnosed at our institution, and on the basis of literature data. A consecutive 3-year series of non-small cell lung cancer (NSCLC) was reviewed according to histological and molecular characteristics. A total of 100 mADCs harboring EGFR exon-19 deletions (61 cases) and the p.(L858R) mutation (39 cases) were selected. All cases were treated by first-line Osimertinib. The prevalence and type of phenotypic transformation were evaluated in patients with available rebiopsy at the time of first-line progression. A total of 32 mADC patients underwent rebiopsy upon first-line Osimertinib progression, and 23 cases had EGFR exon-19 in-frame deletions and 9 p.(L858R) mutations. Four cases showed a phenotypic transformation after a median of 15 months from the start of Osimertinib treatment. All these cases harbored EGFR exon-19 deletions and TP53 pathogenic mutations on diagnostic tumor tissues. Three cases switched to small cell lung cancer histology; in one case, a MET amplification was also detected on rebiopsy. One case changed to spindle cell carcinoma. All cases maintained the initial activating EGFR alteration. For three cases, liquid biopsy was performed at the time of progression: one was negative, one presented only an EGFR exon-19 deletion, and one presented only a MET amplification. In our study, phenotypic transformation had a considerable prevalence among EGFR-positive mADC patients treated by first-line Osimertinib. Different types of histological changes were detected as the only resistance mechanism except for one case with a simultaneously acquired MET amplification. Moreover, all cases harbored TP53 alterations, influencing treatment response. Despite the usefulness of liquid biopsy, rebiopsy should be executed whenever possible. Indeed, it remains the only tool for assessing histological transformation, which greatly impacts prognosis and treatment decisions.

Cluster of Differentiation 47 (CD47), an innate immune checkpoint, facilitates immune escape by binding signal regulatory protein alpha (SIRPα) to inhibit macrophage phagocytosis. Its significance in colorectal cancer (CRC) has garnered heightened interest. This review summarizes five immunohistochemistry (IHC) studies and complementary transcriptomic analyses assessing CD47 in CRC. IHC results consistently indicated membrane overexpression, though positivity rates varied widely (16–91%) due to methodological heterogeneity. Transcriptomic results confirmed CD47 upregulation, especially in Consensus Molecular Subtype 1 (CMS1) and CMS4 subtypes and revealed co-expression with immune checkpoints and oncogenic pathways. Clinically, high CD47 levels were associated with advanced TNM stage, metastasis, poor differentiation, and altered immune infiltration; however, the prognostic significance varied among cohorts. Overall, CD47 appears to be a promising biomarker and therapeutic target, but clinical translation requires standardized evaluation, including harmonized antibody selection and scoring cut-offs, and prospective validation.

In colorectal cancer (CRC), tumor progression is influenced by immunosuppressive tumor microenvironment (TME), in which innate immunity and adaptive immunity play an important role. CD47 is one of the key molecules in the process. It sends a “don’t eat me” signal to macrophages by binding to signal regulatory protein alpha (SIRPα), thus helping tumor cells escape immune clearance. Forkhead box P3 (FOXP3)⁺ regulatory immune cells further suppress antitumor T-cell responses. Here, based on a review of the literature and publicly available transcriptomic data, we propose that CD47 expression and FOXP3⁺ regulatory T cells in CRC are interconnected components of a broader myeloid–regulatory immunosuppressive phenotype, rather than a simple linear CD47–FOXP3 pathway. Evidence from cancer studies and exploratory GEPIA3/TIMER3.0 analyses supports a weak and method-dependent association between CD47 expression, FOXP3 transcripts, and estimated Treg infiltration. Hippo/Yes-associated protein/transcriptional coactivator with PDZ-binding motif (YAP/TAZ) signaling serves as a potential upstream program contributing to this immune context.