Cytotoxic cancer therapy can induce accelerated growth of surviving cancer cells, a phenomenon known as tumor repopulation. This report uncovers a mechanism by which caspase 3 activation in treated cells promotes growth of surviving cells, mediated by iPLA2 and PGE2. The level of caspase 3 activation in human tumors also correlates with risk of relapse, suggesting that this pathway may be a determinant of therapeutic effects. In cancer treatment, apoptosis is a well-recognized cell death mechanism through which cytotoxic agents kill tumor cells. Here we report that dying tumor cells use the apoptotic process to generate potent growth-stimulating signals to stimulate the repopulation of tumors undergoing radiotherapy. Furthermore, activated caspase 3, a key executioner in apoptosis, is involved in the growth stimulation. One downstream effector that caspase 3 regulates is prostaglandin E2 (PGE2), which can potently stimulate growth of surviving tumor cells. Deficiency of caspase 3 either in tumor cells or in tumor stroma caused substantial tumor sensitivity to radiotherapy in xenograft or mouse tumors. In human subjects with cancer, higher amounts of activated caspase 3 in tumor tissues are correlated with markedly increased rate of recurrence and death. We propose the existence of a cell death–induced tumor repopulation pathway in which caspase 3 has a major role.

Purpose: To define human papillomavirus (HPV) –positive oropharyngeal cancers (OPC) suitable for treatment deintensification according to low risk of distant metastasis (DM). Patients and Methods: OPC treated with radiotherapy (RT) or chemoradiotherapy (CRT) from 2001 to 2009 were included. Outcomes were compared for HPV-positive versus HPV-negative patients. Univariate and multivariate analyses identified outcome predictors. Recursive partitioning analysis (RPA) stratified the DM risk. Results: HPV status was ascertained in 505 (56%) of 899 consecutive OPCs. Median follow-up was 3.9 years. HPV-positive patients (n = 382), compared with HPV-negative patients (n = 123), had higher local (94% v 80%, respectively, at 3 years; P < .01) and regional control (95% v 82%, respectively; P < .01) but similar distant control (DC; 90% v 86%, respectively; P = .53). Multivariate analysis identified that HPV negativity (hazard ratio [HR], 2.9; 95% CI, 2.0 to 5.0), N2b-N3 (HR, 2.9; 95% CI, 1.8 to 4.9), T4 (HR, 1.8; 95% CI, 1.2 to 2.9), and RT alone (HR, 1.8; 95% CI, 1.1 to 2.5) predicted a lower recurrence-free survival (RFS; all P < .01). Smoking pack-years > 10 reduced overall survival (HR, 1.72; 95% CI, 1.1 to 2.7; P = .03) but did not impact RFS (HR, 1.1; 95% CI, 0.7 to 1.9; P = .65). RPA segregated HPV-positive patients into low (T1-3N0-2c; DC, 93%) and high DM risk (N3 or T4; DC, 76%) groups and HPV-negative patients into different low (T1-2N0-2c; DC, 93%) and high DM risk (T3-4N3; DC, 72%) groups. The DC rates for HPV-positive, low-risk N0-2a or less than 10 pack-year N2b patients were similar for RT alone and CRT, but the rate was lower in the N2c subset managed by RT alone (73% v 92% for CRT; P = .02). Conclusion: HPV-positive T1-3N0-2c patients have a low DM risk, but N2c patients from this group have a reduced DC when treated with RT alone and seem less suited for deintensification strategies that omit chemotherapy.

Purpose: To refine stage and prognostic group for human papillomavirus (HPV) –related nonmetastatic (M0) oropharyngeal cancer (OPC). Methods: All patients with nonmetastatic (M0) p16-confirmed OPC treated with radiotherapy with or without chemotherapy from 2000 to 2010 were included. Overall survival (OS) was compared among TNM stages for patients with HPV-related and HPV-unrelated OPC separately. For HPV-related OPC, recursive partitioning analysis (RPA) derived new RPA stages objectively. Cox regression was used to calculate adjusted hazard ratios (AHRs) to derive AHR stages. The performance of survival prediction of RPA stage and AHR stage was assessed against the current seventh edition TNM stages. Prognostic groups were derived by RPA, combining RPA stage and nonanatomic factors. Results: The cohort comprised 573 patients with HPV-related OPC and 237 patients with HPV-unrelated OPC, with a median follow-up of 5.1 years. Lower 5-year OS with higher TNM stage was evident for patients with HPV-unrelated OPC (stage I, II, III, and IV 5-year OS: 70%, 58%, 50%, and 30%, respectively; P = .004) but not for patients with HPV-related OPC (stage I, II, III, and IV 5-year OS: 88%, 78%, 71%, and 74%, respectively; P = .56). RPA divided HPV-related OPC into RPA-I (T1-3N0-2b), RPA-II (T1-3N2c), and RPA-III (T4 or N3; 5-year OS: 82%, 76%, and 54%, respectively; P < .001). AHR also yielded a valid classification, but RPA stage demonstrated better survival prediction. A further RPA (including RPA stage, age, and smoking pack-years [PYs]) derived the following four valid prognostic groups for survival: group I (T1-3N0-N2c_≤ 20 PY), group II (T1-3N0-N2c_> 20 PY), group III (T4 or N3_age ≤ 70), and group IVA (T4 or N3_age > 70; 5-year OS: 89%, 64%, 57%, and 40%, respectively; P < .001). Conclusion: An RPA-based TNM stage grouping (stage I/II/III: T1-3N0-N2b/T1-3N2c/T4 or N3, with M1 as stage IV) is proposed for HPV-related OPC as a result of significantly improved survival prediction compared with the seventh edition TNM, and prognostication is further improved by an RPA-based prognostic grouping within the American Joint Committee on Cancer/Union for International Cancer Control TNM framework for HPV-related OPC.