Oncologic surgery is practiced today as it has been for millennia; the surgeon looks for and sees the tumour, and executes a strategy for the most complete and efficient removal. Intraoperatively surgeons still largely rely on inspection and palpation to determine surgical margins in real-time¹. Given this inherently imprecise technique, positive tumour margins are encountered often. For example, a second operation, due to inadequate tumour removal during breast cancer surgery, is required for 20–25% of patients, delaying recovery, causing patient trauma, and adding to healthcare costs as well as worsening patient prognosis². Techniques that can improve the intraoperative visualization of tumour margins would improve the completeness of surgical resection while minimizing the removal of normal tissue.

Surgeons rely heavily on preoperative imaging with modalities such as computed tomography (CT) and magnetic resonance imaging (MRI), but these do not provide real-time information to the surgeon during the operation and do not easily mesh with the optical and tactile methods that surgeons have used since the earliest days of surgery. To integrate imaging into the surgical workflow, optical imaging strategies are being developed using fluorescent probes, targeting tumour markers^1,2. One limitation of the marker-targeting strategy is the lack of broad tumour applicability in cancer patients. For example, although the folate receptor is widely expressed in ovarian cancer, expression in other cancers, such as head-and-neck cancer, is limited. Even within a given type of cancer, expression can be variable; < 25% breast cancer patients have Her2 expression³.

To achieve universal tumour targeting, our group has been working with ubiquitous tumour markers, or “hallmarks of cancer”⁴. We have previously shown that targeting two such hallmarks, angiogenesis and the acidic tumour extracellular microenvironment, allows for imaging of a wide variety of cancers using fluorescence⁵. We have subsequently discovered that the acidic cancer microenvironment, first described by Otto Warburg⁶, can be imaged in a unique way that enhances the information provided to surgeons.

One important reason for the slow adoption of image-guided surgery is that variable target expression results in variable fluorescence. This ambiguity does not provide sufficient additional data to the surgeon to help with the decision of whether to remove tissue. To overcome this ambiguity, we turned to a library of pH sensitive nanoprobes with binary on/off fluorescent responses that are finely tuneable in a broad range of physiological pH (4.0–7.4)⁷. The ultra pH-sensitive property is a unique nanoscale phenomenon arising from a catastrophic phase transition during pH-triggered self-assembly of amphiphilic copolymers⁸. At the molecular level, an all-or-nothing protonation distribution in cationic unimers or neutral micelles, respectively, was observed (Video 1), in contrast to what occurs with conventional polymeric bases. The resulting pH cooperativity results in a dramatically sharpened pH response (∆pH_off/on < 0.2 pH), compared to 2 pH units for small molecular pH sensors. We adopted a nanoprobe with a pKa of 6.9 as a pH threshold sensor to image the acidic tumour microenvironment. This nanoprobe stays completely dark at physiologic pH, and is activated to full fluorescence in a variety of tumours, because most tumours have an average extracellular pH around 6.8. A nearly binary tumour fluorescence was achieved over the muscle background for margin delineation that correlates with the histologic boundary of the tumour (Fig. 1, dashed line). This sharp tumour fluorescence allowed real-time image-guided surgery with significantly improved long-term survival in mice bearing breast and head-and-neck cancers⁹.

Figure 1 | Schematic of pH nanotransistor with binary on/off responses at a threshold pH (pH_t = 6.9). The delineation of tumour margins guided by fluorescence (center) correlates with histology (right).

Surgeons use complex pattern recognition (acquired by surgical experience) to distinguish tumour from normal tissue. The final decision, however, is binary; to remove or not to remove tissue. To help with that decision, an imaging agent must minimize ambiguous signals. The binary response of the nanoprobes achieves this when compared to probes that respond linearly to target concentrations. The demonstration of the value of that binary response is an exciting and tremendously important step for image-guided surgery.

Video 1 | Explanation of the transistor-like effect in the pH nanoprobes.

Our paper: Zhao, T. et al. A transistor-like pH nanoprobe for tumour detection and image-guided surgery. Nat. Biomed. Eng. 1, 0006 (2016).

References

1. Vahrmeijer, A. L., Hutteman, M., van der Vorst, J. R., van de Velde, C. J. & Frangioni, J. V. Image-guided cancer surgery using near-infrared fluorescence. Nat. Rev. Clin. Oncol. 10, 507-518, doi:10.1038/nrclinonc.2013.123 (2013).

2. Nguyen, Q. T. & Tsien, R. Y. Fluorescence-guided surgery with live molecular navigation--a new cutting edge. Nat. Rev. Cancer 13, 653-662, doi:10.1038/nrc3566 (2013).

3. Paik, S. et al. HER2 and choice of adjuvant chemotherapy for invasive breast cancer: national surgical adjuvant breast and bowel project protocol B-15. J. Natl. Cancer Inst. 92, 1991-1998 (2000).

4. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646-674, doi:10.1016/j.cell.2011.02.013 (2011).

5. Wang, Y. et al. A nanoparticle-based strategy for the imaging of a broad range of tumours by nonlinear amplification of microenvironment signals. Nature materials 13, 204-212 (2014).

6. Warburg, O. On the origin of cancer cells. Science 123, 309-314 (1956).

7. Ma, X. et al. Ultra-pH-sensitive nanoprobe library with broad pH tunability and fluorescence emissions. Journal of the American Chemical Society 136, 11085-11092 (2014).

8. Li, Y. et al. Molecular basis of cooperativity in pH-triggered supramolecular self-assembly. Nat Commun 7, 13214, doi:10.1038/ncomms13214 (2016).