Upper gastrointestinal (GI) cancers, including esophageal, gastric, and pancreatic cancers, remain among the most lethal malignancies worldwide, mainly because they resist nearly every therapeutic modality, from platinum-based chemotherapy, and anti-HER2 and anti-VEGF agents, to immune checkpoint inhibitors. Although tumor cell-intrinsic resistance is well characterized, an increasing share of treatment failure traces to the tumor microenvironment (TME), where cancer-associated fibroblasts (CAFs) and the extracellular matrix (ECM) act not as passive stroma but as active orchestrators of resistance. Here we argue that the functional heterogeneity of CAFs, spanning myofibroblastic (myCAF), inflammatory (iCAF), and antigen-presenting (apCAF) subtypes, and the desmoplastic ECM they construct converge on a small number of shared resistance programs. Those include paracrine signaling, metabolic reprogramming, extracellular vesicle (EV) transfer, and biomechanical remodeling that together drive chemoresistance, targeted therapy evasion, and immune exclusion. Emerging spatial multi-omics now resolves these programs to define niches within upper GI tumors, reframing resistance as a spatially organized property of the tissue rather than the tumor cell alone. We bring this evidence together and evaluate strategies aimed at CAF reprogramming and ECM normalization, arguing that spatially resolved targeting of the stroma represents a tractable path to overcoming resistance in these refractory cancers.

Cervical cancer remains one of the most commonly diagnosed cancers in women worldwide, with an elevated number of associated annual deaths, even though effective preventive vaccines are available. The pathophysiology of cervical cancer is well understood, with high-risk HPV as the main culprit in carcinogenesis, in addition to cell proliferation signaling alteration and tumor immune microenvironment modulation. This understanding of the disease’s molecular signatures has led to the development of several lines of treatment, especially for advanced, recurrent, persistent, or metastatic disease. For instance, Bevacizumab, a monoclonal antibody targeting angiogenesis factors, as well as Endostar, a recombinant human endostatin, have been studied and shown to improve survival in advanced disease. In contrast, anti-angiogenic Tyrosine Kinase Inhibitors had limited efficacy. Furthermore, antibody-drug conjugates such as Tisotumab Vedotin allow to deliver a highly toxic payload directly to the tumor site by binding to tissue factor, which is highly expressed in cervical tumor cells. Moreover, immunotherapy has emerged as a key treatment modality in cervical cancer by inhibiting immune checkpoint inhibitors (PD1, PD-L1, and CTLA4). In addition, therapeutic vaccines have been investigated for the treatment of localized disease by enhancing cell-mediated immunity against E6 and E7 proteins. However, more robust clinical trials are needed before these vaccines can be effectively and safely used clinically. Finally, several ongoing trials are currently evaluating new therapeutic modalities and combinations of the currently available tools in the cervical cancer treatment armamentarium.